This action might not be possible to undo. Are you sure you want to continue?
Jay Kant Pratap Singh Yadav
Definition: A DFA is 5tuple or quintuple M = (Q, ¿, o, q
0
, A) where
Q is nonempty, finite set of states.
¿ is nonempty, finite set of input alphabets.
o is transition function, which is a mapping from Q x ¿ to Q.
q
0
e Q is the start state.
A _ Q is set of accepting or final states.
Note: For each input symbol a, from a given state there is exactly one transition (there can be
no transitions from a state also) and we are sure (or can determine) to which state the
machine enters. So, the machine is called Deterministic machine. Since it has finite number
of states the machine is called Deterministic finite machine or Deterministic Finite
Automaton or Finite State Machine (FSM).
The language accepted by DFA is
L(M) = { w  w e ¿* and o*(q
0
, w) e A }
The nonacceptance of the string w by an FA or DFA can be defined in formal notation as:
L(M) = { w  w e ¿* and o*(q
0
, w) e A }
Obtain a DFA to accept strings of a’s and b’s starting with the string ab
Fig.2.7 Transition diagram to accept string ab(a+b)*
So, the DFA which accepts strings of a’s and b’s starting with the string ab is given by M =
(Q, ¿ , o, q
0
, A) where
Q = {q
0
, q
1
, q
2
, q
3
}
¿ = {a, b}
q
0
is the start state
A = {q
2
}.
o is shown the transition table 2.4.
÷E÷
o
a b
÷
S
t
a
t
e
s
÷
÷q
0
q
1
q
3
q
1
q
3
q
2
q
2
q
2
q
3
q
3
q
3
q
2
a,b
q
2
q
1
q
0
a b
q
3
a
a,b
b
2
Draw a DFA to accept string of 0’s and 1’s ending with the string 011.
Obtain a DFA to accept strings of a’s and b’s having a sub string aa
Obtain a DFA to accept strings of a’s and b’s except those containing the substring aab.
Obtain DFAs to accept strings of a’s and b’s having exactly one a,
Obtain a DFA to accept strings of a’s and b’s having even number of a’s and b’s
The machine to accept even number of a’s and b’s is shown in fig.2.22.
q
3
q
1
q
2
q
0
1 0
0
0
1
1 1 0
q
2
q
1
b
q
0
a a
b
a,b
q
3
a b a
q
2
q
1
q
0
a,b
b
b
a
q
1
q
0
b a,b
a
q
2
b
a
q
1
b
q
0
a, b
a
q
1
b
q
0
b
a a
q
3
b
q
2
b
a a
q
4
a, b
q
1
q
0
b
q
2
b
a
a
b
q
3
b
a
a
3
Fig.2.22 DFA to accept even no. of a’s and b’s
Regular language
Definition: Let M = (Q, ¿, o, q
0
, A) be a DFA. The language L is regular if there exists a
machine M such that L = L(M).
* Applications of Finite Automata *
String matching/processing
Compiler Construction
The various compilers such as C/C++, Pascal, Fortran or any other compiler is designed
using the finite automata. The DFAs are extensively used in the building the various phases
of compiler such as
 Lexical analysis (To identify the tokens, identifiers, to strip of the comments etc.)
 Syntax analysis (To check the syntax of each statement or control statement used in
the program)
 Code optimization (To remove the un wanted code)
q
1
q
0
b
q
2
b
a
a
b
q
3
b
a
a
q
1
q
0
b
q
2
b
a
a
b
q
3
b
a
a
q
1
q
0
b
q
2
b
a
a
b
q
3
b
a
a
4
 Code generation (To generate the machine code)
Other applications
The concept of finite automata is used in wide applications. It is not possible to list all the
applications as there are infinite number of applications. This section lists some applications:
1. Large natural vocabularies can be described using finite automaton which includes
the applications such as spelling checkers and advisers, multilanguage dictionaries,
to indent the documents, in calculators to evaluate complex expressions based on the
priority of an operator etc. to name a few. Any editor that we use uses finite
automaton for implementation.
2. Finite automaton is very useful in recognizing difficult problems i.e., sometimes it is
very essential to solve an undecidable problem. Even though there is no general
solution exists for the specified problem, using theory of computation, we can find
the approximate solutions.
3. Finite automaton is very useful in hardware design such as circuit verification, in
design of the hardware board (mother board or any other hardware unit), automatic
traffic signals, radio controlled toys, elevators, automatic sensors, remote sensing or
controller etc.
4. In game theory and games wherein we use some control characters to fight against a
monster, economics, computer graphics, linguistics etc., finite automaton plays a very
important role.
Non deterministic finite automata(NFA)
Definition: An NFA is a 5tuple or quintuple M = (Q, ¿, o, q
0
, A) where
Q is non empty, finite set of states.
¿ is non empty, finite set of input alphabets.
o is transition function which is a mapping from
Q x {¿ U c} to subsets of 2
Q
. This function shows
the change of state from one state to a set of states
based on the input symbol.
q
0
e Q is the start state.
A _ Q is set of final states.
Acceptance of language
Definition: Let M = (Q, ¿, o, q
0
, A) be a DFA where Q is set of finite states, ¿ is set of input
alphabets (from which a string can be formed), o is transition function from Q x {¿Uc} to
2
Q
, q
0
is the start state and A is the final or accepting state. The string (also called language)
w accepted by an NFA can be defined in formal notation as:
L(M) = { w  w e ¿*and o*(q
0
, w) = Q with atleast one
5
Component of Q in A}
Obtain an NFA to accept the following language L = {w  w e abab
n
or aba
n
where n > 0}
The machine to accept either abab
n
or aba
n
where n > 0 is shown below:
2.1 Conversion from NFA to DFA
Let M
N
= (Q
N
, ¿
N
, o
N
, q
0
, A
N
) be an NFA and accepts the language L(M
N
). There should be
an equivalent DFA M
D
= (Q
D
, ¿
D
, o
D
, q
0
, A
D
) such that L(M
D
) = L(M
N
). The procedure to
convert an NFA to its equivalent DFA is shown below:
Step1:
The start state of NFA M
N
is the start state of DFA M
D
. So, add q
0
(which is the start
state of NFA) to Q
D
and find the transitions from this state. The way to obtain
different transitions is shown in step2.
Step2:
For each state [q
i
, q
j
,….q
k
] in Q
D
, the transitions for each input symbol in ¿ can be
obtained as shown below:
1. o
D
([q
i
, q
j
,….q
k
], a) = o
N
(q
i
, a) U o
N
(q
j
, a) U ……o
N
(q
k
, a)
= [q
l
, q
m
,….q
n
] say.
2. Add the state [q
l
, q
m
,….q
n
] to Q
D
, if it is not already in Q
D
.
3. Add the transition from [q
i
, q
j
,….q
k
] to [q
l
, q
m
,….q
n
] on the input symbol a iff the
state [q
l
, q
m
,….q
n
] is added to Q
D
in the previous step.
Step3:
The state [q
a
, q
b
,….q
c
] e Q
D
is the final state, if at least one of the state in q
a
, q
b
, …..
q
c
e A
N
i.e., at least one of the component in [q
a
, q
b
,….q
c
] should be the final state of
NFA.
Step4:
If epsilon (e) is accepted by NFA, then start state q
0
of DFA is made the final state.
Convert the following NFA into an equivalent DFA.
b
q
1
a
q
3
q
2
b a
q
4
q
5
a
q
6
a b
q
7
q
0
c
c
6
¿
D
F
A
Q
D
F
A
Step1: q
0
is the start of DFA (see step1 in the conversion procedure).
So, Q
D
= {[q
0
]} (2.7)
Step2: Find the new states from each state in Q
D
and obtain the corresponding transitions.
Consider the state [q
0
]:
When a = 0
o
D
([q
0
], 0) =
=
o
N
([q
0
], 0)
[q
0
, q
1
]
(2.8)
When a = 1
o
D
([q
0
], 1) =
=
o
N
([q
0
], 1)
[q
1
]
(2.9)
Since the states obtained in (2.8) and (2.9) are not in Q
D
(2.7), add these two states to Q
D
so
that
Q
D
= {[q
0
], [q
0
, q
1
], [q
1
] } (2.10)
The corresponding transitions on a = 0 and a = 1 are shown below.
Consider the state [q
0
, q
1
]:
When a = 0
o
D
([q
0
, q
1
],
0)
=
=
=
=
o
N
([q
0
, q
1
], 0)
o
N
(q
0
, 0) U o
N
(q
1
, 0)
{q
0
, q
1
} U {q
2
}
[q
0
, q
1
, q
2
]
(2.11)
When a = 1
o
D
([q
0
, q
1
], = o
N
([q
0
, q
1
], 1)
o
0 1
[q
0
] [q
0
, q
1
] [q
1
]
[q
0
, q
1
]
[q
1
]
q
1
q
0
0 1
0,1
q
2
0, 1
7
¿
D
F
A
D
F
A
Q
D
F
A
D
F
A
¿
1) =
=
=
o
N
(q
0
, 1) U o
N
(q
1
, 1)
{q
1
} U {q
2
}
[q
1
, q
2
]
(2.12)
Since the states obtained in (2.11) and (2.12) are the not defined in Q
D
(see 2.10), add these
two states to Q
D
so that
Q
D
= {[q
0
], [q
0
, q
1
], [q
1
], [q
0
, q
1
, q
2
], [q
1
, q
2
] } (2.13)
and add the transitions on a = 0 and a = 1 as shown below:
Consider the state [q
1
]:
When a = 0
o
D
([q
1
], 0) =
=
o
N
([q
1
], 0)
[q
2
]
(2.14)
When a = 1
o
D
([q
1
], 1) =
=
o
N
([q
1
], 1)
[q
2
]
(2.15)
Since the states obtained in (2.14) and (2.15) are same and the state q
2
is not in Q
D
(see 2.13),
add the state q
2
to Q
D
so that
Q
D
= {[q
0
], [q
0
, q
1
], [q
1
], [q
0
, q
1
, q
2
], [q
1
, q
2
], [q
2
]} (2.16)
and add the transitions on a = 0 and a = 1 as shown below:
o
0 1
[q
0
] [q
0
, q
1
] [q
1
]
[q
0
, q
1
] [q
0
, q
1
, q
2
] [q
1
, q
2
]
[q
1
]
[q
0
, q
1
,
q
2
]
[q
1
, q
2
]
o
0 1
[q
0
] [q
0
, q
1
] [q
1
]
[q
0
, q
1
] [q
0
, q
1
, q
2
] [q
1
, q
2
]
8
Q
D
F
A
D
F
A
¿
D
F
A
Q
D
Consider the state [q
0
,q
1
,q
2
]:
When a = 0
o
D
([q
0
,q
1
,q
2
],
0)
=
=
=
=
o
N
([q
0
,q
1
,q
2
], 0)
o
N
(q
0
, 0) U o
N
(q
1
, 0) U o
N
(q
2
, 0)
{q
0
,q
1
} U {q
2
} U {}
[q
0
,q
1
,q
2
]
(2.17)
When a = 1
o
D
([q
0
,q
1
,q
2
],
1)
=
=
=
=
o
N
([q
0
,q
1
,q
2
], 1)
o
N
(q
0
, 1) U o
N
(q
1
, 1) U o
N
(q
2
, 1)
{q
1
} U {q
2
} U {q
2
}
[q
1
, q
2
]
(2.18)
Since the states obtained in (2.17) and (2.18) are not new states (are already in Q
D
, see 2.16),
do not add these two states to Q
D
. But, the transitions on a = 0 and a = 1 should be added to
the transitional table as shown below:
Consider the state [q
1
,q
2
]:
When a = 0
o
D
([q
1
,q
2
], 0) = o
N
([q
1
,q
2
], 0)
[q
1
] [q
2
] [q
2
]
[q
0
, q
1
,
q
2
]
[q
1
, q
2
]
[q
2
]
o
0 1
[q
0
] [q
0
, q
1
] [q
1
]
[q
0
, q
1
] [q
0
, q
1
, q
2
] [q
1
, q
2
]
[q
1
] [q
2
] [q
2
]
[q
0
, q
1
,
q
2
]
[q
0
,q
1
,q
2
] [q
1
, q
2
]
[q
1
, q
2
]
[q
2
]
9
Q
D
F
A
¿
D
F
A
D
F
A
=
=
=
o
N
(q
1
, 0) U o
N
(q
2
, 0)
{q
2
} U {}
[q
2
]
(2.19)
When a = 1
o
D
([q
1
,q
2
], 1) =
=
=
=
o
N
([q
1
,q
2
], 1)
o
N
(q
1
, 1) U o
N
(q
2
, 1)
{q
2
} U {q
2
}
[q
2
]
(2.20)
Since the states obtained in (2.19) and (2.20) are not new states (are already in Q
D
see 2.16),
do not add these two states to Q
D
. But, the transitions on a = 0 and a = 1 should be added to
the transitional table as shown below:
Consider the state [q
2
]:
When a = 0
o
D
([q
2
], 0) =
=
o
N
([q
2
], 0)
{}
(2.21)
When a = 1
o
D
([q
2
], 1) =
=
o
N
([q
2
], 1)
[q
2
]
(2.22)
Since the states obtained in (2.21) and (2.22) are not new states (are already in Q
D
, see 2.16),
do not add these two states to Q
D
. But, the transitions on a = 0 and a = 1 should be added to
the transitional table. The final transitional table is shown in table 2.14. and final DFA is
shown in figure 2.35.
o
0 1
[q
0
] [q
0
, q
1
] [q
1
]
[q
0
, q
1
] [q
0
, q
1
, q
2
] [q
1
, q
2
]
[q
1
] [q
2
] [q
2
]
[q
0
, q
1
,
q
2
]
[q
0
,q
1
,q
2
] [q
1
, q
2
]
[q
1
, q
2
] [q
2
] [q
2
]
[q
2
]
10
Fig.2.35 The DFA
Convert the following NFA to its equivalent DFA.
Let Q
D
= {0} (A)
Consider the state [A]:
When input is a:
o
0 1
[q
0
] [q
0
, q
1
] [q
1
]
[q
0
, q
1
, q
2
] [q
1
, q
2
]
[q
2
] [q
2
]
[q
0
,q
1
,q
2
] [q
1
, q
2
]
[q
2
] [q
2
]
[q
2
]

[q
2
]
a
9
4
3
c
6
5
7
c
b
a
e
c
8
2
e c
c
c
1
0
b
[q
2
]
[q 1
]
[q
0
]
[q 0 , q 1
]
[q
1
, q
2
]
[q
0
, q
1
, q
2
]
0
1
0
1
0, 1
0, 1
0
1
1
[q
0
,q
1
]
[q
1
]
[q
1
,q
2
]
[q0q
1
,q
2
]
11
o(A, a) =
=
o
N
(0, a)
{1}
(B)
When input is b:
o( A, b) =
=
o
N
(0, b)
{}
Consider the state [B]:
When input is a:
o(B, a) =
=
o
N
(1, a)
{}
When input is b:
o( B, b) =
=
=
o
N
(1, b)
{2}
{2,3,4,6,9} (C)
This is because, in state 2, due to ctransitions (or without giving any input) there
can be transition to states 3,4,6,9 also. So, all these states are reachable from state 2.
Therefore,
o(B, b) = {2,3,4,6,9} = C
Consider the state [C]:
When input is a:
o(C, a) =
=
=
=
o
N
({2,3,4,6,9}, a)
{5}
{5, 8, 9, 3, 4, 6}
{3, 4, 5, 6, 8, 9} (ascending
order) (D)
This is because, in state 5 due to ctransitions, the states reachable are {8, 9, 3, 4,
6}. Therefore,
o(C, a) = {3, 4, 5, 6, 8, 9} = D
When input is b:
o( C, b) =
=
=
=
o
N
({2, 3, 4, 6, 9}, b)
{7}
{7, 8, 9, 3, 4, 6}
{3, 4, 6, 7, 8, 9}(ascending order)
12
(E)
This is because, from state 7 the states that are reachable without any input (i.e., c
transition) are {8, 9, 3, 4, 6}. Therefore,
o(C, b) = {3, 4, 6, 7, 8, 9} = E
Consider the state [D]:
When input is a:
o(D, a) =
=
=
=
o
N
({3,4,5,6,8,9}, a)
{5}
{5, 8, 9, 3, 4, 6}
{3, 4, 5, 6, 8, 9} (ascending
order) (D)
When input is b:
o(D, b) =
=
=
=
o
N
({3,4,5,6,8,9}, b)
{7}
{7, 8, 9, 3, 4, 6}
{3, 4, 6, 7, 8, 9} (ascending
order) (E)
Consider the state [E]:
When input is a:
o(E, a) =
=
=
=
o
N
({3,4,6,7,8,9}, a)
{5}
{5, 8, 9, 3, 4, 6}
{3, 4, 5, 6, 8, 9}(ascending order)
(D)
When input is b:
o(E, b) =
=
=
=
o
N
({3,4,6,7,8,9}, b)
{7}
{7, 8, 9, 3, 4, 6}
{3, 4, 6, 7, 8, 9}(ascending order)
(E)
Since there are no new states, we can stop at this point and the transition table for the DFA is
shown in table 2.15.
13
Q
D
F
A
D
F
A
¿
D
F
A
D
F
A
Table 2.15 Transitional table
The states C,D and E are final states, since 9 (final state of NFA) is present in C, D and E.
The final transition diagram of DFA is shown in figure 2.36
Fig. 2.36 The DFA
Regular Languages
Regular expression
Definition: A regular expression is recursively defined as follows.
1.  is a regular expression denoting an empty language.
2. c(epsilon) is a regular expression indicates the language containing an empty string.
3. a is a regular expression which indicates the language containing only {a}
4. If R is a regular expression denoting the language L
R
and S is a regular expression
denoting the language L
S
, then
a. R+S is a regular expression corresponding to the language L
R
UL
S
.
o
a b
A B 
B  C
D E
D E
D E
D
E
E
D C B
b a
b
a
A
a
b
b
a
C
14
b. R.S is a regular expression corresponding to the language L
R
.L
S.
.
c. R* is a regular expression corresponding to the language L
R
*
.
5. The expressions obtained by applying any of the rules from 14 are regular
expressions.
The table 3.1 shows some examples of regular expressions and the language corresponding to these
regular expressions.
Regular
expressions
Meaning
(a+b)* Set of strings of a’s and b’s of any length
including the NULL string.
(a+b)*abb Set of strings of a’s and b’s ending with the
string abb
ab(a+b)* Set of strings of a’s and b’s starting with the
string ab.
(a+b)*aa(a+b)
*
Set of strings of a’s and b’s having a sub string
aa.
a*b*c* Set of string consisting of any number of
a’s(may be empty string also) followed by any
number of b’s(may include empty string)
followed by any number of c’s(may include
empty string).
a
+
b
+
c
+
Set of string consisting of at least one ‘a’
followed by string consisting of at least one ‘b’
followed by string consisting of at least one ‘c’.
aa*bb*cc* Set of string consisting of at least one ‘a’
followed by string consisting of at least one ‘b’
followed by string consisting of at least one ‘c’.
(a+b)* (a +
bb)
Set of strings of a’s and b’s ending with either a
or bb
(aa)*(bb)*b Set of strings consisting of even number of a’s
followed by odd number of b’s
(0+1)*000 Set of strings of 0’s and 1’s ending with three
consecutive zeros(or ending with 000)
(11)* Set consisting of even number of 1’s
Table 3.1 Meaning of regular expressions
Obtain a regular expression to accept a language consisting of strings of a’s and b’s of even length.
String of a’s and b’s of even length can be obtained by the combination of the strings aa, ab,
ba and bb. The language may even consist of an empty string denoted by c. So, the regular
expression can be of the form
(aa + ab + ba + bb)*
15
The * closure includes the empty string.
Note: This regular expression can also be represented using set notation as
L(R) = {(aa + ab + ba + bb)
n
 n > 0}
Obtain a regular expression to accept a language consisting of strings of a’s and b’s of odd length.
String of a’s and b’s of odd length can be obtained by the combination of the strings aa, ab,
ba and bb followed by either a or b. So, the regular expression can be of the form
(aa + ab + ba + bb)* (a+b)
String of a’s and b’s of odd length can also be obtained by the combination of the strings aa,
ab, ba and bb preceded by either a or b. So, the regular expression can also be represented as
(a+b) (aa + ab + ba + bb)*
Note: Even though these two expression are seems to be different, the language
corresponding to those two expression is same. So, a variety of regular expressions can be
obtained for a language and all are equivalent.
Obtain NFA from the regular expression
Theorem: Let R be a regular expression. Then there exists a finite automaton M = (Q, ¿, o,
q
0
, A) which accepts L(R).
Proof: By definition, , c and a are regular expressions. So, the corresponding machines to
recognize these expressions are shown in figure 3.1.a, 3.1.b and 3.1.c respectively.
(a) (b) (c)
Fig 3.1 NFAs to accept , c and a
The schematic representation of a regular expression R to accept the language L(R) is shown
in figure 3.2. where q is the start state and f is the final state of machine M.
Fig 3.2 Schematic representation of FA accepting L(R)
In the definition of a regular expression it is clear that if R and S are regular expression, then
R+S and R.S and R* are regular expressions which clearly uses three operators ‘+’, ‘‘ and
q
f
q
0
c
q
f
q
0
a q
f
q
0

q
M
f
L(R)
16
‘.’. Let us take each case separately and construct equivalent machine. Let M
1
= (Q
1
, ¿
1
, o
1
,
q
1
, f
1
) be a machine which accepts the language L(R
1
) corresponding to the regular
expression R
1
. Let M
2
= (Q
2
, ¿
2
, o
2
, q
2
, f
2
) be a machine which accepts the language L(R
2
)
corresponding to the regular expression R
2
.
Case 1: R = R
1
+ R
2
. We can construct an NFA which accepts either L(R
1
) or L(R
2
) which
can be represented as L(R
1
+ R
2
) as shown in figure 3.3.
Fig. 3.3 To accept the language L(R1 + R2)
It is clear from figure 3.3 that the machine can either accept L(R
1
) or L(R
2
). Here, q
0
is the
start state of the combined machine and q
f
is the final state of combined machine M.
Case 2: R = R
1
. R
2
. We can construct an NFA which accepts L(R
1
) followed by L(R
2
) which
can be represented as L(R
1
. R
2
) as shown in figure 3.4.
Fig. 3.4To accept the language L(R1 . R2)
It is clear from figure 3.4 that the machine after accepting L(R
1
) moves from state q
1
to f
1
.
Since there is a ctransition, without any input there will be a transition from state f
1
to state
q
2
. In state q
2
, upon accepting L(R
2
), the machine moves to f
2
which is the final state. Thus,
q
1
which is the start state of machine M
1
becomes the start state of the combined machine M
and f
2
which is the final state of machine M
2
, becomes the final state of machine M and
accepts the language L(R
1
.R
2
).
Case 3: R = (R
1
)
*
. We can construct an NFA which accepts either L(R
1
)
*
) as shown in figure
3.5.a. It can also be represented as shown in figure 3.5.b.
L(R
1
)
q
f
q
1
M
1
f
1
c
q
0
c
c
c
q
f
q
1
M
1
f
1
c
q
0
c
c q
2
M
2
f
2
c
L(R
1
)
L(R
2
)
q
1
M
1
f
1
q
2
M
2
f
2
c
L(R
1
) L(R
2
)
17
(a)
(b)
Fig. 3.5 To accept the language L(R1)
*
It is clear from figure 3.5 that the machine can either accept c or any number of L(R
1
)s thus
accepting the language L(R
1
)
*
. Here, q
0
is the start state q
f
is the final state.
Obtain an NFA which accepts strings of a’s and b’s starting with the string ab.
The regular expression corresponding to this language is ab(a+b)*.
Step 1: The machine to accept ‘a’ is shown below.
Step 2: The machine to accept ‘b’ is shown below.
Step 3: The machine to accept (a + b) is shown below.
Step 4: The machine to accept (a+b)* is shown below.
4
3
c
6
5
7
c
b
a
c
c
8
2
9
4
3
c
6
5
7
c
b
a
c
c
8
c c
c
c
7
6
b
5
4
a
q
1
M
1
f
1
c c
q
0
q
f
c
c
18
Step 5: The machine to accept ab is shown below.
Step 6: The machine to accept ab(a+b)* is shown below.
Fig. 3.6 To accept the language L(ab(a+b)*)
Obtain the regular expression from FA
Theorem: Let M = (Q, ¿, o, q
0
, A) be an FA recognizing the language L. Then there exists
an equivalent regular expression R for the regular language L such that L = L(R).
The general procedure to obtain a regular expression from FA is shown below. Consider the
generalized graph
Fig. 3.9 Generalized transition graph
where r
1
, r
2
, r
3
and r
4
are the regular expressions and correspond to the labels for the edges.
The regular expression for this can take the form:
r = r
1
*
r
2
(r
4
+ r
3
r
1
*
r
2
)
*
(3.1)
Note:
1. Any graph can be reduced to the graph shown in figure 3.9. Then substitute the
regular expressions appropriately in the equation 3.1 and obtain the final regular
expression.
2. If r
3
is not there in figure 3.9, the regular expression can be of the form
r = r
1
*
r
2
r
4
*
(3.2)
3. If q
0
and q
1
are the final states then the regular expression can be of the form
r = r
1
* + r
1
*
r
2
r
4
*
(3.3)
a
9
4
3
c
6
5
7
c
b
a
c
c
8
2
c c
c
c
1
0
b
2
1
0
a b
q
1
q
0
r
2
r
3
r
4
r
1
19
Obtain a regular expression for the FA shown below:
The figure can be reduced as shown below:
It is clear from this figure that the machine accepts strings of 01’s and 10’s of any length and
the regular expression can be of the form
(01 + 10)
*
What is the language accepted by the following FA
Since, state q
2
is the dead state, it can be removed and the following FA is obtained.
The state q
0
is the final state and at this point it can accept any number of 0’s which can be
represented using notation as
0
*
q
1
is also the final state. So, to reach q
1
one can input any number of 0’s followed by 1 and
followed by any number of 1’s and can be represented as
0
*
11
*
0
q
1
q
0
1
q
2
0
1
0
q
3
1
0,1
q
0
01
10
q
1
q
0
1
1 0
0,
1
q
2
0
q
1
q
0
1
1 0
20
So, the final regular expression is obtained by adding 0
*
and 0
*
11
*
. So, the regular expression
is
R.E = 0
*
+ 0
*
11
*
= 0
*
( e + 11
*
)
= 0
*
( e + 1
+
)
= 0
*
(1
*
) = 0
*
1
*
It is clear from the regular expression that language consists of any number of 0’s (possibly
c) followed by any number of 1’s(possibly c).
Applications of Regular Expressions
Pattern Matching refers to a set of objects with some common properties. We can match an
identifier or a decimal number or we can search for a string in the text.
An application of regular expression in UNIX editor ed.
In UNIX operating system, we can use the editor ed to search for a specific pattern in the
text. For example, if the command specified is
/acb*c/
then the editor searches for a string which starts with ac followed by zero or more b’s and
followed by the symbol c. Note that the editor ed accepts the regular expression and searches
for that particular pattern in the text. As the input can vary dynamically, it is challenging to
write programs for string patters of these kinds.
21
Jay Kant Pratap Singh Yadav
Properties of regular languages
• Pumping Lemma
Used to prove certain languages like L = {0
n
1
n
 n ≥ 1} are not regular.
• Closure properties of regular languages
Used to build recognizers for languages that are constructed from other languages by
certain operations.
Ex. Automata for intersection of two regular languages
• Decision properties of regular languages
– Used to find whether two automata define the same language
– Used to minimize the states of DFA
eg. Design of switching circuits.
Pumping Lemma for regular languages ( Explanation)
Let L = {0
n
1
n
 n ≥ 1}
There is no regular expression to define L. 00*11* is not the regular expression defining L.
Let L= {0
2
1
2
}
State 6 is a trap state, state 3 remembers that two 0’s have come and from there state 5
remembers that two 1’s are accepted.
This implies DFA has no memory to remember arbitrary ‘n’. In other words if we have to
remember n, which varies from 1 to ·, we have to have infinite states, which is not possible
with a finite state machine, which has finite number of states.
1 2 3
0
4 5
6
0
1 1
0,1
0
0
1
1
0,1
22
Pumping Lemma (PL) for Regular Languages
Theorem:
Let L be a regular language. Then there exists a constant ‘n’ (which depends
on L) such that for every string w in L such that w ≥ n, we can break w into three strings,
w=xyz, such that:
1. y > 0
2. xy ≤ n
3. For all k ≥ 0, the string xy
k
z is also in L.
PROOF:
Let L be regular defined by an FA having ‘n’ states. Let w= a
1
,a
2
,a
3
a
n
and is in L.
w = n ≥ n. Let the start state be P
1
. Let w = xyz where x= a
1
,a
2
,a
3
a
n1
, y=a
n
and z = c.
Therefore xy
k
z = a
1
 a
n1
(a
n
)
k
c
k=0 a
1
 a
n1
is accepted
k=1 a
1
 a
n
is accepted
k=2 a
1
 a
n+1
is accepted
k=10 a
1
 a
n+9
is accepted and so on.
23
Uses of Pumping Lemma:  This is to be used to show that, certain languages are not regular.
It should never be used to show that some language is regular. If you want to show that
language is regular, write separate expression, DFA or NFA.
General Method of proof: 
(i) Select w such that w > n
(ii) Select y such that y > 1
(iii) Select x such that xy s n
(iv) Assign remaining string to z
(v) Select k suitably to show that, resulting string is not in L.
Example 1.
To prove that L={ww c a
n
b
n
, where n ≥ 1} is not regular
Proof:
Let L be regular. Let n is the constant (PL Definition). Consider a word w in L.
Let w = a
n
b
n
, such that w=2n. Since 2n > n and L is regular it must satisfy PL.
xy contain only a’s. (Because xy ≤ n).
Let y=l, where l > 0 (Because y > 0).
Then, the break up of x. y and z can be as follows
from the definition of PL , w=xy
k
z, where k=0,1,2,·, should belong to L.
That is a
nl
(a
l
)
k
b
n
eL, for all k=0,1,2, ·
Put k=0. we get a
nl
b
n
e L.
Contradiction. Hence the Language is not regular.
Example 2.
To prove that L={ww is a palindrome on {a,b}*} is not regular. i.e., L={aabaa, aba,
abbbba,…}
Proof:
Let L be regular. Let n is the constant (PL Definition). Consider a word w in L. Let w
= a
n
ba
n,
such that w=2n+1. Since 2n+1 > n and L is regular it must satisfy PL.
24
xy contain only a’s. (Because xy ≤ n).
Let y=l, where l > 0 (Because y > 0).
That is, the break up of x. y and z can be as follows
from the definition of PL w=xy
k
z, where k=0,1,2,·, should belong to L.
That is a
nl
(a
l
)
k
ba
n
eL, for all k=0,1,2, ·.
Put k=0. we get a
nl
b a
n
e L, because, it is not a palindrome. Contradiction, hence the
language is not regular
.
Example 3.
To prove that L={ all strings of 1’s whose length is prime} is not regular. i.e., L={1
2
,
1
3
,1
5
,1
7
,1
11
,}
Proof: Let L be regular. Let w = 1
p
where p is prime and  p = n +2
Let y = m.
by PL xy
k
z eL
 xy
k
z  =  xz  +  y
k
 Let k = pm
= (pm) + m (pm)
= (pm) (1+m)  this can not be prime
if pm ≥ 2 or 1+m ≥ 2
1. (1+m) ≥ 2 because m ≥ 1
2. Limiting case p=n+2
(pm) ≥ 2 since m ≤n
Example 4.
To prove that L={ 0
i
2
 i is integer and i >0} is not regular. i.e., L={0
2
, 0
4
,0
9
,0
16
,0
25
,}
Proof: Let L be regular. Let w = 0
n
2
where w = n
2
≥ n
by PL xy
k
z eL, for all k = 0,1,
Select k = 2
 xy
2
z  =  xyz  +  y 
25
= n
2
+ Min 1 and Max n
Therefore n
2
<  xy
2
z  ≤ n
2
+ n
n
2
<  xy
2
z  < n
2
+ n + 1+n adding 1 + n ( Note that less than or equal to is
n
2
<  xy
2
z  < (n + 1)
2
replaced by less than sign)
Say n = 5 this implies that string can have length > 25 and < 36
which is not of the form 0
i
2
.
Exercises for students: 
a) Show that following languages are not regular
(i) L={a
n
b
m
 n, m >0 and n<m }
(ii) L={a
n
b
m
 n, m >0 and n>m }
(iii)L={a
n
b
m
c
m
d
n
 n, m >1 }
(iv) L={a
n
 n is a perfect square }
(v) L={a
n
 n is a perfect cube }
b) Apply pumping lemma to following languages and understand why we cannot complete
proof
(i) L={a
n
aba  n >0 }
(ii) L={a
n
b
m
 n, m >0 }
26
Closure Properties of Regular Languages
1. The union of two regular languages is regular.
2. The intersection of two regular languages is regular.
3. The complement of a regular language is regular.
4. The difference of two regular languages is regular.
5. The reversal of a regular language is regular.
6. The closure (star) of a regular language is regular.
7. The concatenation of regular languages is regular.
8. A homomorphism (substitution of strings for symbols) of a regular language is regular.
9. The inverse homomorphism of a regular language is regular
Closure under Union
Theorem: If L and M are regular languages, then so is L M.
Ex1.
L1={a,a
3
,a
5
,}
L2={a
2
,a
4,
a
6
,}
L1L2 = {a,a
2
,a
3
,a
4
,}
RE=a(a)
*
Ex2.
L1={ab, a
2
b
2
, a
3
b
3
, a
4
b
4
,}
L2={ab,a
3
b
3
,a
5
b
5
,}
L1L2 = {ab,a
2
b
2
, a
3
b
3
, a
4
b
4
, a
5
b
5
}
RE=ab(ab)
*
Closure Under Complementation
Theorem : If L is a regular language over alphabet S, then L = E
*
 L is also a regular
language.
Ex1.
L1={a,a
3
,a
5
,}
E
*
L1={e,a
2
,a
4
,a
6
,}
RE=(aa)*
27
Ex2. Consider a DFA, A that accepts all and only the strings of 0’s and 1’s that end in 01.
That is L(A) = (0+1)
*
01. The complement of L(A) is therefore all string of 0’s and 1’s that do
not end in 01
28
Theorem:  If L is a regular language over alphabet E, then, L = E
*
 L is also
a regular language
Proof:  Let L =L(A) for some DFA. A=(Q, E, o, q
0
, F). Then L = L(B), where B is the
DFA (Q, E, o, q
0
, QF). That is, B is exactly like A, but the accepting states of A have
become nonaccepting states of B, and vice versa, then w is in L(B) if and only if o^ ( q
0,
w)
is in QF, which occurs if and only if w is not in L(A).
Closure Under Intersection
Theorem : If L and M are regular languages, then so is L · M.
Ex1.
L1={a,a
2
,a
3
,a
4
,a
5
,a
6
,}
L2={a
2
,a
4
,a
6
,}
L1L2 = {a
2
,a
4
,a
6
,}
RE=aa(aa)*
Ex2
L1={ab,a
3
b
3
,a
5
b
5
,a
7
b
7
}
L2={a
2
b
2
, a
4
b
4
, a
6
b
6
,}
L1·L2 = 
RE= 
Ex3.
Consider a DFA that accepts all those strings that have a 0.
Consider a DFA that accepts all those strings that have a 1.
The product of above two automata is given below.
29
This automaton accepts the intersection of the first two languages: Those languages that have
both a 0 and a 1. Then pr represents only the initial condition, in which we have seen neither
0 nor 1. Then state qr means that we have seen only once 0’s, while state ps represents the
condition that we have seen only 1’s. The accepting state qs represents the condition where
we have seen both 0’s and 1’s.
Ex 4 (on intersection)
Write a DFA to accept the intersection of L1=(a+b)*a and L2=(a+b)*b that is for L1 · L2.
DFA for L1 · L2 =  (as no string has reached to final state (2,4))
Ex5 (on intersection)
Find the DFA to accept the intersection of L1=(a+b)*ab (a+b)* and L2=(a+b)*ba
(a+b)* that is for L1 · L2
DFA for L1 · L2
30
Closure Under Difference
Theorem : If L and M are regular languages, then so is L – M.
Ex.
L1={a,a
3
,a
5
,a
7
,}
L2={a
2
,a
4
,a
6
,}
L1L2 = {a,a
3
,a
5
,a
7
}
RE=a(a)
*
Reversal
Theorem : If L is a regular language, so is L
R
Ex.
L={001,10,111,01}
L
R
={100,01,111,10}
To prove that regular languages are closed under reversal.
Let L = {001, 10, 111}, be a language over E={0,1}.
L
R
is a language consisting of the reversals of the strings of L.
That is L
R
= {100,01,111}.
If L is regular we can show that L
R
is also regular.
Proof.
As L is regular it can be defined by an FA, M = (Q, E , o, q
0
, F), having only one final state.
If there are more than one final states, we can use e transitions from the final states going
to a common final state.
31
Let FA, M
R
= (Q
R
, E
R
, o
R
,q
0
R
,F
R
) defines the language L
R
,
Where Q
R
= Q, E
R
= E, q
0
R
=F,F
R
=q
0
, and o
R
(p,a)> q, iff o (q,a) > p
Since M
R
is derivable from M, L
R
is also regular.
The proof implies the following method
1. Reverse all the transitions.
2. Swap initial and final states.
3. Create a new start state p0 with transition on e to all the
accepting states of original DFA
Example
Let r=(a+b)* ab define a language L. That is
L = {ab, aab, bab,aaab, }. The FA is as given below
The FA for L
R
can be derived from FA for L by swapping initial and final states and
changing the direction of each edge. It is shown in the following figure.
32
Jay Kant Pratap Singh Yadav
Homomorphism
A string homomorphism is a function on strings that works by substituting a particular string
for each symbol.
Theorem : If L is a regular language over alphabet E, and h is a homomorphism on E, then h
(L) is also regular.
Ex.
The function h defined by h(0)=ab h(1)=c is a homomorphism.
h applied to the string 00110 is ababccab
L1= (a+b)* a (a+b)*
h : {a, b} {0, 1}*
Resulting :
h1(L) = (01 + 11)* 01 (01 + 11)*
h2(L) = (101 + 010)* 101 (101 + 010)*
h3(L) = (01 + 101)* 01 (01 + 101)*
Inverse Homomorphism
Theorem : If h is a homomorphism from alphabet S to alphabet T, and L is a regular
language over T, then h
1 (
L) is also a regular language.
33
Ex.Let L be the language of regular expression (00+1)
*
.
Let h be the homomorphism defined by h(a)=01 and h(b)=10. Then h
1
(L) is the language of
regular expression (ba)
*
.
Decision Properties of Regular Languages
1. Is the language described empty?
2. Is a particular string w in the described language?
3. Do two descriptions of a language actually describe the same language?
This question is often called “equivalence” of languages.
Converting Among Representations
Converting NFA’s to DFA’s
Time taken for either an NFA or NFA to DFA can be exponential in the number of states of
the NFA. Computing cClosure of n states takes O(n
3
) time. Computation of DFA takes
O(n
3
) time where number of states of DFA can be 2
n
. The running time of NFA to DFA
conversion including c transition is O(n
3
2
n
). Therefore the bound on the running time is
O(n
3
s) where s is the number of states the DFA actually has.
DFA to NFA Conversion
Conversion takes O(n) time for an n state DFA.
Automaton to Regular Expression Conversion
For DFA where n is the number of states, conversion takes O(n
3
4
n
) by substitution method
and by state elimination method conversion takes O(n
3
) time. If we convert an NFA to DFA
and then convert the DFA to a regular expression it takes the time O(n
3
4
n
3
2
n
)
Regular Expression to Automaton Conversion
Regular expression to cNFA takes linear time – O(n) on a regular expression of length n.
Conversion from cNFA to NFA takes O(n
3
) time.
Testing Emptiness of Regular Languages
Suppose R is regular expression, then
1. R = R1 + R2. Then L(R) is empty if and only if both L(R1) and L(R2) are empty.
2. R= R1R2. Then L(R) is empty if and only if either L(R1) or L(R2) is empty.
3. R=R1* Then L(R) is not empty. It always includes at least c
4. R=(R1) Then L(R) is empty if and only if L(R1) is empty since they are the same
language.
Testing Emptiness of Regular Languages
Suppose R is regular expression, then
1. R = R1 + R2. Then L(R) is empty if and only if both L(R1) and L(R2) are empty.
34
2. R= R1R2. Then L(R) is empty if and only if either L(R1) or L(R2) is empty.
3. R=(R1)* Then L(R) is not empty. It always includes at least c
4. R=(R1) Then L(R) is empty if and only if L(R1) is empty since they are the same
language.
Testing Membership in a Regular Language
Given a string w and a Regular Language L, is w in L.
If L is represented by a DFA, simulate the DFA processing the string of input symbol
w, beginning in start state. If DFA ends in accepting state the answer is ‘Yes’ , else it is ‘no’.
This test takes O(n) time
If the representation is NFA, if w is of length n, NFA has s states, running time of this
algorithm is O(ns
2
)
If the representation is c  NFA, c  closure has to be computed, then processing of
each input symbol , a , has 2 stages, each of which requires O(s
2
) time.
If the representation of L is a Regular Expression of size s, we can convert to an c 
NFA with almost 2s states, in O(s) time. Simulation of the above takes O(ns
2
) time on an
input w of length n
Minimization of Automata ( Method 1)
Let p and q are two states in DFA. Our goal is to understand when p and q (p ≠ q) can
be replaced by a single state.
Two states p and q are said to be distinguishable, if there is at least one string, w, such
that one of o^ (p,w) and o^ (q,w) is accepting and the other is not accepting.
Algorithm 1:
List all unordered pair of states (p,q) for which p ≠ q. Make a sequence of passes
through these pairs. On first pass, mark each pair of which exactly one element is in F. On
each subsequent pass, mark any pair (r,s) if there is an ae¿ for which o (r,a) = p, o (s,a) = q,
and (p,q) is already marked. After a pass in which no new pairs are marked, stop. The
marked pair (p,q) are distinguishable.
Examples:
1. Let L = {e, a
2
, a
4
, a
6
, ….} be a regular language over ¿ = {a,b}. The FA is shown
in Fig 1.
Fig 2. gives the list of all unordered pairs of states (p,q) with p ≠ q.
35
The boxes (1,2) and (2,3) are marked in the first pass according to the algorithm 1.
In pass 2 no boxes are marked because, o(1,a)  and o (3,a) 2. That is (1,3) (,2),
where  and 3 are non final states.
o(1,b)  and o (3,b) . That is (1,3) (,), where  is a nonfinal state. This implies
that (1,3) are equivalent and can replaced by a single state A.
Fig 3. Minimal Automata corresponding to FA in Fig 1
Minimization of Automata (Method 2)
Consider set {1,3}. (1,3) (2,2) and (1,3) (,). This implies state 1 and 3 are
equivalent and can not be divided further. This gives us two states 2,A. The resultant FA is
shown is Fig 3.
Example 2. (Method1):
Let r= (0+1)*10, then L(r) = {10,010,00010,110, }. The FA is given below
Following fig shows all unordered pairs (p,q) with p ≠ q
36
The pairs marked 1 are those of which exactly one element is in F; They are marked on pass
1. The pairs marked 2 are those marked on the second pass. For example (5,2) is one of these,
since (5,2) (6,4), and the pair (6,4) was marked on pass 1.
From this we can make out that 1, 2, and 4 can be replaced by a single state 124 and
states 3, 5, and 7 can be replaced by the single state 357. The resultant minimal FA is shown
in Fig. 6
The transitions of fig 4 are mapped to fig 6 as shown below
Example 2. (Method1):
(2,3) (4,6) this implies that 2 and 3 belongs to different group hence they are split in level
2. similarly it can be easily shown for the pairs (4,5) (1,7) and (2,5) and so on.
37
Jay Kant Pratap Singh Yadav
Lecture Notes for Class 4 (21/4/2005)
Context Free Grammar
Context Free grammar or CGF, G is represented by four components that is G= (V, T, P, S),
where V is the set of variables, T the terminals, P the set of productions and S the start
symbol.
Example: The grammar G
pal
for palindromes is represented by
G
pal
= ({P}, {0, 1}, A, P)
Where A represents the set of five productions
1. Pe
2. P0
3. P1
4. P0P0
5. P1P1
Derivation using Grammar
Example 1: Leftmost Derivation
The inference that a * (a+b00) is in the language of variable E can be reflected in a derivation
of that string, starting with the string E. Here is one such derivation:
E E * E I * E a * E
a * (E) a * (E + E) a * (I + E) a * (a + E)
a * (a + I) a * (a + I0) a * (a + I00) a * (a + b00)
38
Leftmost Derivation  Tree
Example 2: Rightmost Derivations
The derivation of Example 1 was actually a leftmost derivation. Thus, we can describe the
same derivation by:
E E * E E *(E) E * (E + E)
E * (E + I) E * (E +I0) E * (E + I00) E * (E + b00)
E * (I + b00) E * (a +b00) I * (a + b00) a * (a + b00)
We can also summarize the leftmost derivation by saying
E a * (a + b00), or express several steps of the derivation by expressions such as E *
E a * (E).
Rightmost Derivation  Tree
There is a rightmost derivation that uses the same replacements for each variable, although it
makes the replacements in different order. This rightmost derivation is:
39
E E * E E * (E) E * (E + E)
E * (E + I) E * (E + I0) E * (E + I00) E * (E + b00)
E * (I + b00) E * (a + b00) I * (a + b00) a * (a + b00)
This derivation allows us to conclude E a * (a + b00)
Consider the Grammar for string(a+b)*c
EE + T  T
T T * F  F
F ( E )  a  b  c
Leftmost Derivation
ETT*FF*F(E)*F(E+T)*F(T+T)*F(F+T)*F (a+T)*F (a+F)*F
(a+b)*F(a+b)*c
Rightmost derivation
ETT*FT*cF*c(E)*c(E+T)*c(E+F)*c
(E+b)*c(T+b)*c(F+b)*c(a+b)*c
Example 2:
Consider the Grammar for string (a,a)
S>(L)a
L>L,SS
Leftmost derivation
S(L)(L,S)(S,S)(a,S)(a,a)
Rightmost Derivation
S(L)(L,S)(L,a)(S,a)(a,a)
The Language of a Grammar
If G(V,T,P,S) is a CFG, the language of G, denoted by L(G), is the set of terminal
strings that have derivations from the start symbol.
L(G) = {w in T  S w}
Sentential Forms
Derivations from the start symbol produce strings that have a special role called “sentential
forms”. That is if G = (V, T, P, S) is a CFG, then any string in (V T)* such that S o is a
sentential form. If S o, then is a left – sentential form, and if S o , then is a right –
sentential form. Note that the language L(G) is those sentential
forms that are in T*; that is they consist solely of terminals.
For example, E * (I + E) is a sentential form, since there is a derivation
E E * E E * (E) E * (E + E) E * (I + E)
40
However this derivation is neither leftmost nor rightmost, since at the last step, the middle E
is replaced.
As an example of a left – sentential form, consider a * E, with the leftmost derivation.
E E * E I * E a * E
Additionally, the derivation
E E * E E * (E) E * (E + E)
Shows that
E * (E + E) is a right – sentential form.
Ambiguity
A context – free grammar G is said to be ambiguous if there exists some w eL(G) which has
at least two distinct derivation trees. Alternatively, ambiguity implies the existence of two or
more left most or rightmost derivations.
Ex:
Consider the grammar G=(V,T,E,P) with V={E,I}, T={a,b,c,+,*,(,)}, and productions.
EI,
EE+E,
EE*E,
E(E),
Iabc
Consider two derivation trees for a + b * c.
41
Now unambiguous grammar for the above
Example:
ET, TF, FI, EE+T, TT*F,
F(E), Iabc
Inherent Ambiguity
A CFL L is said to be inherently ambiguous if all its grammars are ambiguous
Example:
Condider the Grammar for string aabbccdd
SAB  C
A aAb  ab
BcBd  cd
C aCd  aDd
D>bDc  bc
Parse tree for string aabbccdd
42
Applications of Context – Free Grammars
• Parsers
• The YACC Parser Generator
• Markup Languages
• XML and Document typr definitions
The YACC Parser Generator
EI  E+E  E*E  (E)
Ia  b  Ia  Ib  I0  I1
Exp : Id {…}
 Exp ‘+’ Exp {…}
 Exp ‘*’ Exp {…}
 ‘(‘ Exp ‘)’ {…}
;
Id : ‘a’ {…}
’b’ {…}
 Id ‘a’ {…}
Id ‘b’ {…}
Id ‘0’ {…}
Id ‘1’ {…}
;
43
XML and Document type definitions.
1. AE1,E2.
ABC
BE1
CE2
2. AE1  E2.
AE1
AE2
3. A(E1)*
ABA
Ac
BE1
4. A(E1)+
ABA
AB
BE1
5. A(E1)?
Ac
AE1
EXERCISE QUESTIONS
1) Design contextfree grammar for the following cases
a) L={ 0n1n  n≥l }
b) L={aibjck i≠j or j≠k}
2) The following grammar generates the language of RE
0*1(0+1)*
S AB
A 0Ac
B 0B1Bc
Give leftmost and rightmost derivations of the following strings
a) 00101 b) 1001 c) 00011
44
3) Consider the grammar
S aSaSbSc
Show that deviation for the string aab is ambiguous
4) Suppose h is the homomorphism from the alphabet {0,1,2} to the alphabet { a,b} defined
by h(0) = a; h(1) = ab &
h(2) = ba
a) What is h(0120) ?
b) What is h(21120) ?
c) If L is the language L(01*2), what is h(L) ?
d) If L is the language L(0+12), what is h(L) ?
e) If L is the language L(a(ba)*) , what is h1(L) ?
45
%{ #include <stdio.h>
%}
%token ID id
%%
Exp : id { $$ = $1 ; printf ("result is %d\n", $1);}
 Exp ‘+’ Exp {$$ = $1 + $3;}
 Exp ‘*’ Exp {$$ = $1 * $3; }
 ‘(‘ Exp ‘)’ {$$ = $2; }
;
%%
int main (void) {
return yyparse ( );
}
void yyerror (char *s) {
fprintf (stderr, "%s\n", s);
}
%{
#include "y.tab.h"
%}
%%
[09]+ {yylval.ID = atoi(yytext); return id;}
[ \t \n] ;
[+ * ( )] {return yytext[0];}
. {ECHO; yyerror ("unexpected character");}
%%
Example 2:
%{
#include <stdio.h>
%}
%start line
%token <a_number> number
%type <a_number> exp term factor
%%
line : exp ';' {printf ("result is %d\n", $1);}
;
exp : term {$$ = $1;}
 exp '+' term {$$ = $1 + $3;}
 exp '' term {$$ = $1  $3;}
term : factor {$$ = $1;}
 term '*' factor {$$ = $1 * $3;}
 term '/' factor {$$ = $1 / $3;}
;
46
factor : number {$$ = $1;}
 '(' exp ')' {$$ = $2;}
;
%%
int main (void) {
return yyparse ( );
}
void yyerror (char *s) {
fprintf (stderr, "%s\n", s);
}
%{
#include "y.tab.h"
%}
%%
[09]+ {yylval.a_number = atoi(yytext); return number;}
[ \t\n] ;
[+*/();] {return yytext[0];}
. {ECHO; yyerror ("unexpected character");}
%%
Markup Languages
Functions
•Creating links between documents
•Describing the format of the document
Example
The Things I hate
1. Moldy bread
2. People who drive too slow
In the fast lane
HTML Source
<P> The things I <EM>hate</EM>:
<OL>
<LI> Moldy bread
<LI>People who drive too slow
In the fast lane
</OL>
47
HTML Grammar
•Char a  A  …
•Text e  Char Text
•Doc e  Element Doc
•Element Text 
<EM> Doc </EM>
<p> Doc 
<OL> List </OL> …
5. ListItem <LI> Doc
6. List e  ListItem List Start symbol
XML and Document type definitions.
2. AE1,E2.
ABC
BE1
CE2
3. AE1  E2.
AE1
AE2
4. A(E1)*
ABA
Ac
BE1
4. A(E1)+
ABA
AB
BE1
5. A(E1)?
Ac
AE1
EXERCISE QUESTIONS
1) Design contextfree grammar for the following cases
a) L={ 0n1n  n≥l }
b) L={aibjck i≠j or j≠k}
3) The following grammar generates the language of RE
0*1(0+1)*
S AB
A 0Ac
B 0B1Bc
Give leftmost and rightmost derivations of the following strings
48
a) 00101 b) 1001 c) 00011
3) Consider the grammar
S aSaSbSc
Show that deviation for the string aab is ambiguous
4) Suppose h is the homomorphism from the alphabet {0,1,2} to the alphabet { a,b} defined
by h(0) = a; h(1) = ab &
h(2) = ba
a) What is h(0120) ?
b) What is h(21120) ?
c) If L is the language L(01*2), what is h(L) ?
d) If L is the language L(0+12), what is h(L) ?
e) If L is the language L(a(ba)*) , what is h1(L) ?
49
Session by Jay K. Pratap Singh Yadav:
Introduction to Push Down Automata
Just as finitestate automata correspond to regular languages, the contextfree languages
(CFL) have corresponding machines called pushdown automata (PDA).
Regular expressions are generators for regular languages and Finite Automata’s are
recognizers for them. Similarly for Contextfree Languages, Context Free Grammars (CFG)
are generators and Pushdown Automata (PDA) are recognizers.
PDA is more powerful than FA. An FA cannot recognize the language a
n
b
n
, n> 0, because
FA does not have any memory to remember the number of a’s it has already seen, for
equating with number of b’s found. PDA is NFA with an added memory. Stack functions as
the required memory. So, a PDA is an NFA with a stack. Figure 1 shows a diagrammatic
representation of PDA. The Finite State Control (FSC) reads inputs, one symbol at a time.
Based on the input symbol, current state and the top symbol on the stack, FSC does some
state transitions and does some operations to the stack content. Stack could be kept
unchanged, or some thing could be pushed into the stack and could be popped out of the
stack.
Formal Definition:
A nondeterministic pushdown automaton or npda is a 7tuple M = (Q, E, I, o, q
0
, z, F),
where
Q is a finite set of states,
E is a the input alphabet,
I is the stack alphabet,
o is a transition function, has the form
o: Q X (E {e}) X I ÷ finite subsets of Q X I*
q
0
eQ is the initial state,
z e I is the stack start symbol, and
F _ Q is a set of final states.
stack
Finite State Control
Input Tape
stack
Finite State Control
stack
Finite State Control
Input Tape
Figure 1: PushDown Automaton
50
Transition function: for any given state, input symbol and stack symbol, gives a new state
and stack symbol; i.e. it has the form: (P, a, t) ÷ (Q, u)
Basically, if, a e E, t e I and P and Q are states. Means “read the symbol ‘a’ from the input,
move from state P to state Q, and replace the symbol ‘t’ on top of the stack with the symbol
‘u’ ”.
If u = t, then stack is unchanged.
If u = e, then stack is popped
If u= wx, then t is replaced with x and w is pushed into the stack.
Example 1:
Construct PDA for the language L= {a
n
b
n
 a, b e Σ n > 0}.
Start at state q
0
and keep Z
0
in the stack. The following transitions are possible:
1. If current state is q
0
, and symbol on input tape is at c, and stack top is Z
0
, then move to q
2
the final state.
2. If current state is q
0
, and input tape symbol is a, and stack top Z
0
, then stay in q
0
and push
‘a’ to the stack.
3. If current state is q
0
, input tape symbol is ‘a’, and stack top is a, stay in q
0
and push ‘a’ to
the stack.
4. If current state is q
0
, input tape symbol is b, stack top is a, move to state q
1
and pop the
top symbol of the stack.
5. If current state is q
1
, input tape symbol is b, stack top is a, stay in q
1
and pop the top
symbol of the stack
6. If current state is q
1
, input tape symbol is c and stack top is Z
0
, move to q
2
the final state.
So we can define PDA as M = ({q
0
,
q
1
,
q
2
}, {a, b}, {a, b, Z
0
}, δ, q
0
, Z
0
, {q
2
}), where δ is
defined by following rules:
δ(q
0
, a, Z
0
) = {(q
0
, aZ
0
)}
δ(q
0
, a, a) = {(q
0
, aa)}
δ(q
0
, b, a) = {(q
1
, c)}
δ(q
1
, b, a) = {(q
1
, c)}
δ(q
1
, c, Z
0
) = {(q
2
, c)}
δ(q
0
, c, Z
0
) = {(q
2
, c)}
δ(q,x,Y) =  for all other possibilities
51
Graphical Notation of PDA:
To understand the behavior or PDA clearer, the transition diagram of PDA can be used.
Transition diagram of PDA is generalization of transition diagram of FA.
1. Node corresponds to states of PDA
2. Arrow labeled Start indicates start state
3. Doubly circled states are final states
4. Arc corresponds to transitions of PDA. If δ(q, a, X) = {(p, o)…} is an arc labeled (a,
X/o) from state q to state p means that an input tape head positioned at symbol a and
stack top with X, moves automaton to state q and replaces the stack top with o.
The transition diagram for the above example PDA is given in Figure 2.
Instantaneous Description:
Instantaneous Description or configuration of a PDA describes its execution status at
any time. Instantaneous Description is a represented by a triplet (q, w, u), where
1. q is the current state of the automaton,
2. w is the unread part of the input string, w e Σ*
3. u is the stack contents, written as a string, with the leftmost symbol at the top of the stack.
So u e Γ
*
Moves of A PDA:
Let the symbol "" indicates a move of the nPDA. There are two types of moves possible
for a PDA.
1. Move by consuming input symbol
q
2
q
0
q
1
b,a/ c
b,a/ c
c, Z
0
/ c
c, Z
0
/ c
a, Z
0
/aZ
0
a,a/aa
q
2
q
0
q
1
b,a/ c
b,a/ c
c, Z
0
/ c
c, Z
0
/ c
a, Z
0
/aZ
0
a,a/aa
Figure 2: Transition diagram
Start
52
Suppose that o(q
1
, a, x) = {(q
2
, y), ...}. Then the following move by consuming an input
symbol is possible:
(q
1
, aW, xZ)  (q
2
, W, yZ),
where W indicates the rest of the input string following the a, and Z indicates the rest of the
stack contents underneath the x. This notation says that in moving from state q
1
to state q
2
, an
input symbol ‘a’ is consumed from the input string aW, and the symbol ‘x’ at the top (left) of
the stack xZ is replaced with symbol ‘y’, leaving yZ on the stack.
The above example PDA with a few example input strings, the moves are given below:
a) Moves for the input string aabb:
(q
0
, aabb, Z
0
)  (q
0
, abb, aZ
0
) as per transition rule δ(q
0
, a, Z
0
) = {(q
0
, aZ
0
)}
 (q
0
, bb, aaZ
0
) as per transition rule δ(q
0
, a, a) = {(q
0
, aa)}
 (q
1
, b, aZ
0
) as per transition rule δ(q
0
, b, a) = {(q
1
, c)}
 (q
1
, c, Z
0
) as per transition rule δ(q
0
, b, a) = {(q
1
, c)}
 (q
2
, c,c) as per transition rule δ(q
1
, c, Z
0
) = {(q
2
, c)}
PDA reached a configuration of (q
2
, c,c). The input tape is empty, stack is empty and PDA
has reached a final state. So the string is accepted.
b) Moves for the input string aaabb:
(q
0
, aaabb, Z
0
)  (q
0
, aabb, aZ
0
) as per transition rule δ(q
0
, a, Z
0
) = {(q
0
, aZ
0
)}
 (q
0
, abb, aaZ
0
) as per transition rule δ(q
0
, a, a) = {(q
0
, aa)}
 (q
0
, bb, aaaZ
0
) as per transition rule δ(q
0
, a, a) = {(q
0
, aa)}
 (q
1
, b, aaZ
0
) as per transition rule δ(q
0
, b, a) = {(q
1
, c)}
 (q
1
, c, aZ
0
) as per transition rule δ(q
0
, b, a) = {(q
1
, c)}
 There is no defined move.
So the automaton stops and the string is not accepted.
c) Moves for the input string aabbb:
(q
0
, aabbb, Z
0
)  (q
0
, abbb, aZ
0
) as per transition rule δ(q
0
, a, Z
0
) = {(q
0
, aZ
0
)}
 (q
0
, bbb, aaZ
0
) as per transition rule δ(q
0
, a, a) = {(q
0
, aa)}
 (q
1
, bb, aZ
0
) as per transition rule δ(q
0
, b, a) = {(q
1
, c)}
 (q
1
, b, Z
0
) as per transition rule δ(q
0
, b, a) = {(q
1
, c)}
53
 There is no defined move.
So the automaton stops and the string is not accepted.
2. c move
Suppose that o(q
1
, c, x) = {(q
2
, y), ...}. Then the following move without consuming an input
symbol is possible:
(q
1
, aW, xZ)  (q
2
, aW, yZ),
This notation says that in moving from state q
1
to state q
2
, an input symbol ‘a’ is not
consumed from the input string aW, and the symbol ‘x’ at the top (left) of the stack xZ is
replaced with symbol ‘y’, leaving yZ on the stack. In this move, the tape head position will
not move forward. This move is usually used to represent nondeterminism.
The relation 
*
is the reflexivetransitive closure of  used to represent zero or more moves
of PDA. For the above example, (q
0
, aabb, Z
0
) * (q
2
, c,c).
Example 2:
Design a PDA to accept the set of all strings of 0’s and 1’s such that no prefix has more 1’s
than 0’s.
Solution: The transition diagram could be given as figure 3.
M = ({a, b, c, d}, {0,1}, {0,1,Z}, o, a, Z, {d}), where o is given by:
o (a, 0, Z) ={(b, 0Z)}
o (b, 0, 0) ={(b, 00)}
o (b, 1, 0) ={(c, c)}
o (c, 0, 0) ={(b, 00)}
o (c, 1, 0) ={(c, c)}
o (b, c, 0) ={(d, 0)}
o (b, c, Z) ={(d, Z)}
o (c, 0, Z) ={(b, 0Z)}
Figure 3: Transition diagram
54
o (c, c, 0) ={(d, 0)}
o (c, c, Z) ={(d, Z)}
For all other moves, PDA stops.
Moves for the input string 0010110 is given by:
(a, 0010110, Z)  (b, 010110, 0Z)  (b,10110, 00Z)  (c, 0110, 0Z)  (b, 110, 00Z) 
(c, 10, 0Z)  (c, 0, Z)  (b, c, 0Z)  (d, c, 0Z).
So the input string is accepted by the PDA.
Moves for 011
(a,011,Z)  (b,11,0Z)  (c,1,Z) no move, so PDA stops
Exercises:
Construct PDA:
1. For the language L = {wcw
R
 w e {a, b}*, c e Σ }
2. Accepting the set of all strings over {a, b} with equal number of a’s and b’s. Show the
moves for abbaba
3. For the language L = {a
n
b
2n
 a, b e Σ, n > 0}
4. Accepting the language of balanced parentheses, (consider any type of parentheses)
55
Session by Jay K. Pratap Singh:
Finite Automata & Formal Languages
Push Down Automata
Languages of PDA
1. Languages of PDA
There are 2 ways of accepting an input string PDA
a. Accept by Final state
After consuming the input, PDA enters a final state. The content of the stack is
irrelevant.
b. Accept by empty stack
After consuming the input, stack of the PDA will be empty. The current state could
be final or nonfinal state.
Both methods are equivalent. It is possible to covert a PDA accept by final state to another
PDA accept by empty stack and also the vice versa. Usually the languages that a PDA accept
by final state and PDA by empty stack are different. For example the language {L = a
n
b
m
 n
> m}, the appropriate PDA could be by final state. After consuming the input, the stack may
not be empty.
Accept by Final state:
Let P = (Q, Σ, Γ, δ, q
0
, Z
0
, F) be a PDA. Then L(P), the language accepted by P by the final
state is
{w  (q
0
, w, Z
0
) * (q, c, o)}, where q e F and o e Γ*
Example: L = {ww
R
 w is in (0 + 1)*}, the language for even length palindrome.
Acceptable input strings are like 00, 1111, 0110, 101101, and 110011. In the string 0110, the
difficulty is how to decide the middle of the input string? The 3rd 1 can be part of w or can
be part of w
R
. The PDA could be constructed as below.
M = ({q
0
,
q
1
,
q
2
}, {0, 1}, {0, 1,Z
0
}, δ, q
0
, Z
0
, q
2
), where δ is defined by:
δ(q
0
, 0, Z
0
) = {(q
0
, 0Z
0
)}
δ(q
0
, 1, Z
0
) = {(q
0
, 1Z
0
)}
δ(q
0
, 0, 0) = {(q
0
, 00)}
δ(q
0
, 1, 1) = {(q
0
, 11)}
δ(q
0
, 0, 1) = {(q
0
, 01)}
δ(q
0
, 1, 0) = {(q
0
, 10)}
δ(q
0
, c, Z
0
) = {(q
1
, Z
0
)}
56
δ(q
0
, c, 0) = {(q
1
, 0)}
δ(q
0
, c, 1) = {(q
1
, 1)}
δ(q
1
, 0, 0) = {(q
1
, c)}
δ(q
1
, 1, 1) = {(q
1
, c)}
δ(q
1
, c, Z
0
) = {(q
2
, Z
0
)}
(q
0
, ww
R
, Z
0
) * (q
0
, w
R
, w
R
Z
0
)  (q
1
, w
R
, w
R
Z
0
) * (q
1
, c, Z
0
)  (q
2
, c, Z
0
)
The transition diagram for the PDA is given in figure 1.
The moves of the PDA for the input string 101101 are given figure 2.
q
2
q
0
q
1
0,0/ c
1,1/ c
c,0/0
c,1/1
c, Z
0
/ Z
0
c, Z
0
/ Z
0
0, Z
0
/0Z
0
1, Z
0
/1Z
0
0,0/00
1,1/11
0,1/ 01
1,0/ 10
q
2
q
0
q
1
0,0/ c
1,1/ c
c,0/0
c,1/1
c, Z
0
/ Z
0
c, Z
0
/ Z
0
0, Z
0
/0Z
0
1, Z
0
/1Z
0
0,0/00
1,1/11
0,1/ 01
1,0/ 10
Figure 1: Transition Diagram for L = {ww
R}
Figure 2: moves of PDA for string 101101
57
Accept by empty stack:
Let PDA P = (Q, Σ, Γ, δ, q
0
, Z
0
). We define the language accepted by empty stack by
N(P) = {w  (q
0
, w, Z
0
) * (q, c, c)}, where q e Q
Example:
Construct PDA to accept by empty stack for the language L = {ww
R
 w is in (0 + 1)*}
Instead of the transition δ(q
1
, c, Z
0
) = {(q
2
, Z
0
)} give δ(q
1
, c, Z
0
) = {(q
2
, c)} to get accept
by empty stack. The set of accepting states are irrelevant. This example also shows that L(P)
= N(P)
Example:
Construct PDA to accept by empty stack for the language L={0
i
1
j
 0 s i s j}
The transition diagram for the PDA is given in Figure 3.
2. Conversion between the two forms:
a. From Empty Stack to Final State:
Theorem: If L = N(P
N
) for some PDA P
N
= (Q, Σ, Γ , δ
N
, q
0
, Z
0
), then there is a PDA P
F
such that L = L(P
F
)
Proof:
The method of conversion is given in figure 4.
p
0
P
f
c, X
0
/Z
0
X
0
q
0
c, X
0
/c
(add this transition from all states of P
N
to
new state P
f
)
P
N
p
0
P
f
c, X
0
/Z
0
X
0
q
0
c, X
0
/c
(add this transition from all states of P
N
to
new state P
f
)
P
N
Figure 4: P
F
simulates P
N
s
q
0, 0/00
0, Z
0
/0Z
0
p
1, 0/c
1,Z
0
/Z
0
1, 0/c
1, Z
0
/Z
0
c, Z
0
/Z
0
1, Z
0
/Z
0
c, Z
0
/Z
0
r
c,Z
0
/c
s
q
0, 0/00
0, Z
0
/0Z
0
p
1, 0/c
1,Z
0
/Z
0
1, 0/c
1, Z
0
/Z
0
c, Z
0
/Z
0
1, Z
0
/Z
0
c, Z
0
/Z
0
r
c,Z
0
/c
Figure 3: transition diagram of 0
i
1
j
 0 s i s j
58
We use a new symbol X
0
, which must be not symbol of Γ to denote the stack start symbol
for P
F.
Also add a new start state p
0
and final state p
f
for P
F.
Let P
F
= (Q{p
0
, p
f
}, Σ,
Γ{X
0
}, δ
F
, p
0
, X
0
, {P
f
}), where δ
F
is defined by
δ
F
(p
0
, c, X
0
) = {(q
0
, Z
0
X
0
)} to push X
0
to the bottom of the stack
δ
F
(q, a, y) = δ
N
(q, a, y) a e Σ or a = c and y e Γ, same for both P
N
and P
F.
δ
F
(q, c, X
0
) = {(P
f
, c)} to accept the string by moving to final state.
The moves of P
F
to accept a string w can be written like:
(p
0
, w, X
0
) 
P
F
(p
0
, w, Z
0
X
0
) *
P
F
(q, c, X
0
)  (P
f
, c, c )
b. From Final State to Empty Stack:
Theorem: If L = L(P
F
) for some PDA P
F
= (Q, Σ, Γ, δ
F
, q
0
, Z
0
, F), then there is a PDA P
N
such that L = N(P
N
)
Proof:
The method of conversion is given in figure 5.
To avoid P
F
accidentally empting its stack, initially change the stack start content from Z
0
to
Z
0
X
0
. Also add a new start state p
0
and final state p for P
N.
Let P
N
= (Q{p
0
, p}, Σ,
Γ{X
0
}, δ
N
, p
0
, X
0
), where δ
N
is defined by:
δ
N
(p
0
, c, X
0
) = {(q
0
, Z
0
X
0
)} to change the stack content initially
δ
N
(q
, a, y) = δ
F
(q
, a, y), a e Σ or a = c and y e Γ, same for both
δ
N
(q
, c, y) = {(p
, c)}, q e F, y e Γ or y = X
0
, same for both
δ
N
(p
, c, y) = {(p, c)}, y e Γ or y = X
0,
to pop the remaining stack contents.
.
The moves of P
N
to accept a string w can be written like:
(p
0
, w, X
0
) 
P
N
(q
0
, w, Z
0
X
0
) *
P
N
(q, c, X
0
)  (p, c, c )
q
0
p
0
p
c, X
0
/Z
0
X
0
c, I/c
P
F
c, I/c
c, I/c
q
0
p
0
p
c, X
0
/Z
0
X
0
c, I/c
P
F
c, I/c
c, I/c
Figure 5: P
N
simulates P
F
59
Example:
Construct PDA to accept by final state the language of all strings of 0’s and 1’s such that
number of 1’s is less than number of 0’s. Also convert the PDA to accept by empty stack.
Solution:
PDA by final state is given byM = ({q
0
,
q
1
}, {0, 1}, {0, 1,Z}, δ, q
0
, Z, {q
1
}), where δ is
given by:
δ(q
0
, 0, Z) = {(q
0
, 0Z)}
δ(q
0
, 1, Z) = {(q
0
, 1Z)}
δ(q
0
, 0, 0) = {(q
0
, 00)}
δ(q
0
, 0,1) = {(q
0
, c)}
δ(q
0
, 1, 1) = {(q
0
, 11)}
δ(q
0
, 1,0) = {(q
0
, c)}
δ(q
0
, c,Z) = {(q
1
, Z)}
δ(q
0
, c,0) = {(q
1
, 0)}
For all other moves, PDA stops.
PDA by empty stack is given by M = ({q
0
,
q
1
,
q
2}
, {0, 1}, {0, 1, Z}, δ’, q
0
, Z), where δ’ is
the union of δ and the transitions given below:
δ(q
1
, c,Z) = {(q
2
, c)}
δ(q
1
, c,0) = {(q
2
, c)}
δ(q
2
, c,0) = {(q
2
, c)}
δ(q
2
, c,Z) = {(q
2
, c)}
Exercises:
Design nPDA to accept the language:
1. {a
i
b
j
c
k
 i, j, k > 0 and i = j or i = k}
2. {a
i
b
j
c
i+j
 i, j > 0}
3. {a
i
b
i+j
c
j
 i > 0, j > 1}
60
Finite Automata & Formal Languages
Push Down Automata
Equivalence of PDA and CFG
I. Equivalence of PDA and CFG
The aim is to prove that the following three classes of languages are same:
1. Context Free Language defined by CFG
2. Language accepted by PDA by final state
3. Language accepted by PDA by empty stack
It is possible to convert between any 3 classes. The representation is shown in figure 1.
From CFG to PDA:
Given a CFG G, we construct a PDA P that simulates the leftmost derivations of G. The
stack symbols of the new PDA contain all the terminal and nonterminals of the CFG. There
is only one state in the new PDA; all the rest of the information is encoded in the stack. Most
transitions are on c, one for each production. New transitions are added, each one
corresponding to terminals of G. For every intermediate sentential form uAo in the leftmost
derivation of w (initially w = uv for some v), M will have Ao on its stack after reading u. At
the end (case u = w) the stack will be empty.
Let G = (V, T, Q, S) be a CFG. The PDA which accepts L(G) by empty stack is given by:
P = ({q}, T, V T, δ, q, S) where δ is defined by:
1. For each variable A include a transition,
δ(q, c, A) = {(q, b)  A ÷ b is a production of Q}
2. For each terminal a, include a transition
δ(q, a, a) = {(q, c)}
CFG
PDA by
empty stack
PDA by
Final state
CFG
PDA by
empty stack
PDA by
Final state
Figure 1: Equivalence of PDA and CFG
61
CFG to PDA conversion is another way of constructing PDA. First construct CFG, and then
convert CFG to PDA.
Example:
Convert the grammar with following production to PDA accepted by empty stack:
S ÷ 0S1  A
A ÷ 1A0  S  c
Solution:
P = ({q}, {0, 1}, {0, 1, A, S}, δ, q, S), where δ is given by:
δ(q, c, S) = {(q, 0S1), (q, A)}
δ(q, c, A) = {(q, 1A0), (q, S), (q, c)}
δ(q, 0, 0) = {(q, c)}
δ(q, 1, 1) = {(q, c)}
From PDA to CFG:
Let P = (Q, Σ, Γ, δ, q
0
, Z
0
) be a PDA. An equivalent CFG is G = (V, Σ, R, S), where
V = {S, [pXq]}, where p, q e Q and X e Γ, productions of R consists of
1. For all states p, G has productions S ÷ [q
0
Z
0
p]
2. Let δ(q,a,X) = {(r, Y
1
Y
2
…Y
k
)} where a e Σ or a = c, k can be 0 or any number and
r
1
r
2
…r
k
are list of states. G has productions
[qXr
k
] ÷ a[rY
1
r
1
] [r
1
Y
2
r
2
] … [r
k1
Y
k
r
k
]
If k = 0 then [qXr] ÷a
Example:
Construct PDA to accept ifelse of a C program and convert it to CFG. (This does not accept
if –if –elseelse statements).
Let the PDA P = ({q}, {i, e}, {X,Z}, δ, q, Z), where δ is given by:
δ(q, i, Z) = {(q, XZ)}, δ(q, e, X) = {(q, c)} and δ(q, c, Z) = {(q, c)}
Solution:
Equivalent productions are:
S ÷ [qZq]
62
[qZq] ÷ i[qXq][qZq]
[qXq] ÷ e
[qZq] ÷ c
If [qZq] is renamed to A and [qXq] is renamed to B, then the CFG can be defined by:
G = ({S, A, B}, {i, e}, {S÷A, A÷iBA  c, B÷ e}, S)
Example:
Convert PDA to CFG. PDA is given by P = ({p,q}, {0,1}, {X,Z}, δ, q, Z)), Transition
function δ is defined by:
δ(q, 1, Z) = {(q, XZ)}
δ(q, 1, X) = {(q, XX)}
δ(q, c, X) = {(q, c)}
δ(q, 0, X) = {(p, X)}
δ(p, 1, X) = {(p, c)}
δ(p, 0, Z) = {(q, Z)}
Solution:
Add productions for start variable
S ÷ [qZq]  [qZp]
For δ(q, 1, Z)= {(q, XZ)}
[qZq] ÷ 1[qXq][qZq]
[qZq] ÷ 1[qXp][pZq]
[qZp] ÷ 1[qXq][qZp]
[qZp] ÷ 1[qXp][pZp]
For δ(q, 1, X)= {(q, XX)}
[qXq] ÷ 1[qXq][qXq]
[qXq] ÷ 1[qXp][pXq]
[qXp] ÷ 1[qXq][qXp]
[qXp] ÷ 1[qXp][pXp]
For δ(q, c, X) = {(q, c)}
[qXq] ÷ c
For δ(q, 0, X) = {(p, X)}
[qXq] ÷ 0[pXq]
[qXp] ÷ 0[pXp]
63
For δ(p, 1, X) = {(p, c)}
[pXp] ÷ 1
For δ(p, 0, Z) = {(q, Z)}
[pZq] ÷ 0[qZq]
[pZp] ÷ 0[qZp]
Renaming the variables [qZq] to A, [qZp] to B, [pZq] to C, [pZp] to D, [qXq] to E [qXp] to
F, [pXp] to G and [pXq] to H, the equivalent CFG can be defined by:
G = ({S, A, B, C, D, E, F, G, H}, {0,1}, R, S). The productions of R also are to be
renamed accordingly.
Exercises:
a. Convert to PDA, CFG with productions:
1. A ÷ aAA, A ÷ aS  bS  a
2. S ÷ SS  (S)  c
3. S ÷ aAS  bAB  aB, A ÷ bBB  aS  a, B ÷ bA  a
b. Convert to CFG, PDA with transition function:
δ(q, 0, Z) = {(q, XZ)}
δ(q, 0, X) = {(q, XX)}
δ(q, 1, X) = {(q, X)}
δ(q, c, X) = {(p, c)}
δ(p, 1, X) = {(p, XX)}
δ(p, c, X) = {(p, c)}
δ(p, 1, Z) = {(p, c)}
II. Deterministic PDA:
NPDA provides nondeterminism to PDA. Deterministic PDA’s (DPDA) are very useful for use in programming languages. For example
Parsers used in YACC are DPDA’s.
Definition:
A PDA P= (Q, Σ, Γ, δ, q
0
, Z
0
, F) is deterministic if and only if,
1.δ(q,a,X) has at most one member for qeQ, a e Σ or a= c and XeΓ
2.If δ(q,a,X) is not empty for some aeΣ, then δ(q, c,X) must be empty
64
DPDA is less powerful than nPDA. The Context Free Languages could be recognized by
nPDA. The class of language DPDA accept is in between than of Regular language and
CFL. NPDA can be constructed for accepting language of palindromes, but not by DPDA.
Example:
Construct DPDA which accepts the language L = {wcw
R
 w e {a, b}*, c e Σ}.
The transition diagram for the DPDA is given in figure 2.
Exercises:
Construct DPDA for the following:
1. Accepting the language of balanced parentheses. (Consider any type of parentheses)
2. Accepting strings with number of a’s is more than number of b’s
3. Accepting {0
n
1
m
 n > m}
DPDA and Regular Languages:
The class of languages DPDA accepts is in between regular languages and CFLs. The
DPDA languages include all regular languages. The two modes of acceptance are not same
for DPDA.
To accept with final state:
If L is a regular language, L=L(P) for some DPDA P. PDA surely includes a stack, but the
DPDA used to simulate a regular language does not use the stack. The stack is inactive
always. If A is the FA for accepting the language L, then δ
P
(q,a,Z)={(p,Z)} for all p, q e Q
such that δ
A
(q,a)=p.
q
2
q
0
c,0/0
c,1/1
c, Z
0
/ Z
0
c, Z
0
/ Z
0
q
1
0,0/ c
1,1/ c 0, Z
0
/0Z
0
1, Z
0
/1Z
0
0,0/00
1,1/11
0,1/ 01
1,0/ 10
q
2
q
0
c,0/0
c,1/1
c, Z
0
/ Z
0
c, Z
0
/ Z
0
q
1
0,0/ c
1,1/ c 0, Z
0
/0Z
0
1, Z
0
/1Z
0
0,0/00
1,1/11
0,1/ 01
1,0/ 10
q
0
c,0/0
c,1/1
c, Z
0
/ Z
0
c, Z
0
/ Z
0
q
1
0,0/ c
1,1/ c 0, Z
0
/0Z
0
1, Z
0
/1Z
0
0,0/00
1,1/11
0,1/ 01
1,0/ 10
c, Z
0
/ Z
0
q
1
0,0/ c
1,1/ c 0, Z
0
/0Z
0
1, Z
0
/1Z
0
0,0/00
1,1/11
0,1/ 01
1,0/ 10
q
1
0,0/ c
1,1/ c 0, Z
0
/0Z
0
1, Z
0
/1Z
0
0,0/00
1,1/11
0,1/ 01
1,0/ 10
0, Z
0
/0Z
0
1, Z
0
/1Z
0
0,0/00
1,1/11
0,1/ 01
1,0/ 10
Figure 2: DPDA L = {wcw
R}
65
To accept with empty stack:
Every regular language is not N(P) for some DPDA P. A language L = N(P) for some DPDA
P if and only if L has prefix property. Definition of prefix property of L states that if x, y e
L, then x should not be a prefix of y, or vice versa. NonRegular language L=WcW
R
could
be accepted by DPDA with empty stack, because if you take any x, ye L(WcW
R
), x and y
satisfy the prefix property. But the language, L={0*} could be accepted by DPDA with final
state, but not with empty stack, because strings of this language do not satisfy the prefix
property. So N(P) are properly included in CFL L, ie. N(P) _ L
DPDA and Ambiguous grammar:
DPDA is very important to design of programming languages because languages DPDA
accept are unambiguous grammars. But all unambiguous grammars are not accepted by
DPDA. For example S ÷ 0S01S1 c is an unambiguous grammar corresponds to the
language of palindromes. This is language is accepted by only nPDA. If L = N(P) for DPDA
P, then surely L has unambiguous CFG.
If L = L(P) for DPDA P, then L has unambiguous CFG. To convert L(P) to N(P) to have
prefix property by adding an end marker $ to strings of L. Then convert N(P) to CFG G’.
From G’ we have to construct G to accept L by getting rid of $ .So add a new production
$÷c as a variable of G.
66
Simplification of CFG
The goal is to take an arbitrary Context Free Grammar G = (V, T, P, S) and perform
transformations on the grammar that preserve the language generated by the grammar but
reach a specific format for the productions. A CFG can be simplified by eliminating
1. Useless symbols
Those variables or terminals that do not appear in any derivation of a terminal string
starting from Start variable, S.
2. c  Productions
A ÷ c, where A is a variable
3. Unit production
A ÷ B, A and B are variables
1. Eliminate useless symbols:
Definition: Symbol X is useful for a grammar G = (V, T, P, S) if there is S
*
¬ oX
*
¬w,
weI*. If X is not useful, then it is useless.
Omitting useless symbols from a grammar does not change the language generated
Example:
S ÷ aSb  c  A, A ÷ aA. Here A is a useless symbol
S÷A, A÷aA  c, B ÷ bB. Here B is a useless symbol
Symbol X is useful if both the conditions given below are satisfied in that order itself.
1. X is generating variable. ie if X *¬ w, where weI*
2. X is reachable. ie if S *¬ oX
Theorem:
Let G = (V, T, P, S) be a CFG and assume that L(G) = , then G
1
=(V
1
, T
1
, P
1
, S) be a
grammar without useless symbols by
1. Eliminating non generating symbols
2. Eliminating symbols that are non reachable
67
Elimination has to be performed only in the order of 1 followed by 2. Otherwise the grammar
produced will not be completely useless symbols eliminated.
Example:
Consider a grammar with productions: S ÷ ABa, A ÷ a. If the order of eliminations is 1
followed by 2 gives S ÷ a. If the order of eliminations is 2 followed by 1 gives S ÷ a, A
÷a. Here A is still useless symbol, so completely useless symbols are not eliminated.
Eliminate nongenerating symbols:
Generating symbols follow to one of the categories below:
1. Every symbol of T is generating
2. If A ÷ o and o is already generating, then A is generating
Nongenerating symbols = V generating symbols.
Example:
G= ({S, A, B}, {a}, S ÷ AB  a, A ÷a}, S)
Solution:
B is a nongenerating symbol. After eliminating B, we get
G1= ({S, A}, {a}, {S ÷ a, A ÷a}, S)
Example:
Eliminate the nongenerating symbols from S ÷ aS  A  C, A ÷a, B ÷ aa, C ÷ aCb
Solution:
C is a nongenerating symbol. After eliminating C gets, S ÷ aS  A, A ÷a, B ÷ aa
Eliminate symbols that are non reachable:
Dependency Graph: For the production C ÷ xDy, the dependency graph is given below.
C D
68
Draw dependency graph for all productions. If there is no edge reaching a variable X from
Start symbol S, then X is non reachable.
Example:
Eliminate nonreachable symbols from G= ({S, A}, {a}, {S ÷ a, A ÷a}, S)
Solution:
Draw the dependency graph as given above. A is nonreachable from S. After eliminating A,
G1= ({S}, {a}, {S ÷ a}, S)
Example:
Eliminate nonreachable symbols from S ÷ aS  A, A ÷ a, B ÷ aa
Draw the dependency graph as given above. B is nonreachable from S. After eliminating B,
we get the grammar with productions S ÷ aS  A, A ÷ a
Example:
Eliminate useless symbols from the grammar with productions S ÷ AB  CA, B ÷BC  AB,
A ÷a, C ÷ AB  b
Step 1: Eliminate nongenerating symbols
V
1
= {A, C, S}
P
1
= {S ÷ CA, A ÷a, C ÷ b}
Step 2: Eliminate symbols that are non reachable
S A
S
A B
S A
C
69
Draw the dependency graph as given above. All Variables are reachable. So the final
variables and productions are same V
1
and P
1
.
V
2
= {A, C, S}
P
2
= {S ÷ CA, A ÷a, C ÷ b}
Exercises:
Eliminate useless symbols from the grammar
1. P= {S ÷ aAa, A ÷Sb  bCC, C ÷abb, E ÷ aC}
2. P= {S ÷ aBa  BC, A ÷ aC  BCC, C ÷a, B ÷ bcc, D ÷ E, E ÷d}
3. P= {S ÷ aAa, A ÷ bBB, B ÷ ab, C ÷ aB}
4. P= {S ÷ aS  AB, A ÷ bA, B ÷ AA}
III. Eliminate c  Productions:
Most theorems and methods about grammars G assume L(G) does not contain c. So if c is not
there in L(G), then we have to find out an equivalent G without cproductions. Example for a
grammar G with c  productions is
S ÷ ABA, A ÷ aA  c, B ÷ bB  c
The procedure to find out an equivalent G with out cproductions
1. Find nullable variables
2. Add productions with nullable variables removed.
3. Remove cproductions and duplicates
Step 1: Find set of nullable variables
Nullable variables: Variables that can be replaced by null (c). If A
*¬
c then A is a
nullable variable.
In the grammar with productions S ÷ ABA, A ÷ aA  c, B ÷ bB  c, A is nullable because
of the production A ÷ c. B is nullable because of the production B ÷ c. S is nullable
because both A and B are nullable.
70
Algorithm to find nullable variables is given below:
V: set of variables
N
0
÷{A  A in V, production A ÷ c}
Repeat
N
i
÷ N
i1
U {A A in V, A ÷ α, α in N
i1
}
until N
i
= N
i1
Step 2: Add productions with nullable variables removed
For each production of the form A ÷ w, create all possible productions of the form A ÷ w’,
where w’ is obtained from w by removing one or more occurrences of nullable variables.
Example:
In the grammar with productions S ÷ ABA, A ÷ aA  c, B ÷ bB  c, A, B and S are nullable
variables. So after removing nullable variables we get the productions
S ÷ ABA  BA  AA  AB  A  B  c
A ÷ aA  c  a
B ÷ bB  c  b
Step 3: Remove cproductions and duplicates
The desired grammar consists of the original productions together with the productions
constructed in step 2, minus any productions of the form A ÷ c.
Example:
For the above example we get the final grammar with productions
S ÷ ABA  BA  AA  AB  A  B
A ÷ aA  a
B ÷ bB  b
Example:
Find out the grammar without c  Productions
G = ({S, A, B, D}, {a}, {S ÷ aS  AB, A ÷ c, B÷ c, D ÷b}, S)
71
Solution:
Nullable variables = {S, A, B}
New Set of productions:
S ÷ aS  a
S ÷ AB  A  B
D ÷ b
G
1
= ({S, B, D}, {a}, {S ÷ aS  a  AB  A  B, D ÷b}, S)
Exercise:
Eliminate c  productions from the grammar
1. S ÷ a Xb  aYa, X ÷ Y c, Y ÷ b  X
2. S ÷ Xa, X ÷ aX  bX  c
3. S ÷ XY, X ÷Zb, Y ÷ bW, Z ÷AB, W ÷Z, A ÷ aA  bB  c, B ÷ Ba  Bb c
4. S ÷ ASB  c, A ÷ aAS  a, B ÷ SbS  A bb
But if you have to get a grammar without c  productions and useless symbols, follow the
sequence given below:
1.
Eliminate c  productions and obtain G
1
2.
Eliminate useless symbols from G
1
and obtain G
2
Example:
Eliminate c  productions and useless symbols from the grammar
S ÷a aABC, A ÷aB c, B ÷aA, C ÷aCD, D ÷ddd
Step 1: Eliminate c  productions
Nullable ={A}
P1={S ÷a aA  B  C, A ÷aB, B ÷aAa, C ÷ aCD, D ÷ ddd
Step 2: Eliminating useless symbols
Step 2a: Eliminate nongenerating symbols
72
Generating ={D, B, A, S}
P2={S ÷a  aA B, A ÷ aB, B ÷ aAa, D ÷ ddd}
Step 2b: Eliminate non reachable symbols
C is nonreachable, eliminating C gets
P3= S ÷a aAB, A ÷ aB, B ÷aAa}
S
A
B D
73
I. Eliminate unit productions:
Definition:
Unit production is of form A ÷ B, where A and B are variables.
Unit productions could complicate certain proofs and they also introduce extra steps into
derivations that technically need not be there. The algorithm for eliminating unit productions
from the set of production P is given below:
1. Add all non unit productions of P to P1
2. For each unit production A ÷ B, add to P1 all productions A ÷ o, where B ÷ o is a
nonunit production in P.
3. Delete all the unit productions
Example:
Eliminate unit productions from S ÷ ABA  BA  AA  AB  A  B, A ÷ aA  a, B ÷ bBb
Solution:
The unit productions are S ÷ A, S ÷B. So A and B are derivable. Now add productions
from derivable.
S÷ ABA  BA  AA  AB  A  B  aA  a  bB  b
A ÷ aA  a
B ÷ bB  b
Remove unit productions from above productions to get
S÷ ABA  BA  AA  AB  aA  a  bB  b
A ÷ aA  a
B ÷ bB  b
Example:
Eliminate unit productions from S ÷ Aa  B, A ÷ a  bc  B, B ÷A  bb
Solution:
74
Unit productions are S ÷ B, A ÷ B and B ÷ A. So A and B are derivable. Add productions
from derivable and eliminate unit productions to get,
S ÷ Aa  bb  a  bc
A ÷ a  bc  bb
B ÷ bb  a  bc
Simplified Grammar:
If you have to get a grammar without c  productions, useless symbols and unit productions,
follow the sequence given below:
1.
Eliminate c  productions from G and obtain G
1
2.
Eliminate unit productions from G
1
and obtain G
2
3.
Eliminate useless symbols from G
2
and obtain G
3
Example:
Eliminate useless symbols, c productions and unit productions from the grammar with
productions:
S ÷ a  aA  B C, A ÷ aB  c, B ÷ aA, C ÷ cCD, D ÷ ddd
Step 1: Eliminate c productions
Nullable = {A}
P
1
= {S ÷ a  aA  B  C, A ÷ aB, B ÷ aA  a, C ÷ cCD, D ÷ ddd}
Step 2: Eliminate unit productions
Unit productions are S ÷ B and S÷ C. So Derivable variables are B and C.
P
2
= {S ÷ a  aA  cCD, A ÷ aB, B ÷ aA  a, C ÷ cCD, D ÷ddd}
Step 3: Eliminate useless symbols
After eliminate nongenerating symbol C we get
P
3
= {S ÷ a  aA, A ÷ aB, B ÷ aA  a, D ÷ ddd}
After eliminate symbols that are non reachable
75
S A B D S A B S A B D
P
4
= {S ÷ a  aA, A ÷ aB, B ÷ aA  a}
So the equivalent grammar G
1
= ({S, A, B}, {a}, {S ÷ a  aA, A ÷ aB, B ÷ aA  a}, S)
II. Chomsky Normal Form (CNF)
Every nonempty CFL without c, has a grammar in CNF with productions of the form:
1. A ÷ BC, where A, B, C e V
2. A ÷ a, where A e V and a e T
Algorithm to produce a grammar in CNF:
1. Eliminate useless symbols, c productions and unit productions from the grammar
2. Elimination of terminals on RHS of a production
a)
Add all productions of the form A ÷ BC or A ÷ a to P
1
b)
Consider a production A ÷ X
1
X
2
…X
n
with some terminals of RHS. If X
i
is
a terminal say a
i
, then add a new variable C
ai
to V
1
and a new production C
ai
÷ a
i
to P
1
. Replace X
i
in A production of P by C
ai
c)
Consider A ÷ X
1
X
2
…X
n,
where n >3 and all X
i
‘s are variables. Introduce
new productions A ÷ X
1
C
1
, C
1
÷ X
2
C
2
, … , C
n2
÷X
n1
X
n
to P
1
and
C
1
, C
2
, … , C
n2
to V
1
Example:
Convert to CNF: S ÷ aAD, A ÷aB  bAB, B ÷b, D ÷d
Solution:
Step1: Simplify the grammar
Grammar is already simplified.
Step2a: Elimination of terminals on RHS
Change S ÷ aAD to S ÷ C
a
AD, C
a
÷ a
A ÷aB to A ÷ C
a
B
A ÷ bAB to A ÷ C
b
AB, C
b
÷ b
76
Step2b: Reduce RHS with 2 variables
Change S ÷ C
a
AD to S ÷ C
a
C
1
, C
1
÷ AD
A ÷ C
b
AB to A ÷ C
b
C
2
, C
2
÷ AB
Grammar converted to CNF is given below:
G
1
=({S, A, B, D, C
a
, C
b
, C
1
, C
2
}, {a, b}, S ÷ C
a
C
1
,
A ÷ C
a
B C
b
C
2
, C
a
÷a, C
b
÷b, C
1
÷AD, C
2
÷AB}, S)
Example:
Convert the grammar with following productions to CNF:
P={S ÷ ASB  c, A ÷ aAS  a, B ÷SbS  A  bb}
Solution:
Step1: Simplify the grammar
Step 1a: Eliminate c productions: Consists of S ÷ c
Eliminating c productions from P to get:
P
1
={S ÷ASBAB, A ÷ aAS  aA  a, B÷ SbS  Sb  bS  b  A  bb}
Step 1b: Eliminate unit productions: B÷ A
P
2
= {S ÷ ASBAB, A ÷aASaAa, B ÷SbS  Sb  bS  b  bb  aAS  aA  a}
Step 1c: Eliminate useless symbols: no useless symbols
Step2: Convert to CNF
P
3
={S ÷ AC
1
 AB, A ÷ C
a
C
2
 C
a
A  a,
B ÷ SC
3
 SC
b
 C
b
S  b  C
b
C
b
 C
a
C
2
 C
a
A  a,
C
a
÷ a, C
b
÷ b, C
1
÷ SB, C
2
÷ AS, C
3
÷ C
b
S}
Exercises:
Convert the following grammar with productions to CNF:
1. S ÷ aSa  bSb  a  b  aa  bb
77
2. S ÷ bA  aB, A ÷ bAA  aS  a, B ÷ aBB  bS  b
3. S÷ Aba, A ÷ aab, B ÷ AC
4. S ÷ 0A0 1B1  BB, A ÷ C, B ÷ SA, C ÷ S c
5. S ÷ aAa  bBb c, A ÷ Ca, B ÷ C  b, C ÷ CDE c, D ÷ A  B  ab
III. Greibach Normal Form (GNF):
Every nonempty language without c is L(G) for some grammar G with productions are of the
form
A ÷ ao,
where a e T and o is a string of zero or more variables. Conversion to GNF is a complex
process. Converting CFG in GNF to PDA gets a PDA without crules.
Example:
Transform into Greibach normal form the grammar with productions
S ÷ 0S1  01
Solution:
S ÷ 0SA  0A, A ÷ 1
Usage of GNF:
Construction of PDA from a GNF grammar can be made more meaningful with GNF.
Assume that the PDA has two states: start state q
0
and accepting state q
1.
An S rule of the
form S ÷ aA
1
A
2
…A
n
generates a transition that processes the terminal a, pushes the
variables A
1
A
2
…A
n
on the stack and enters q
1.
The remainder of the computation uses the
input symbol and the stack top to determine the appropriate transition.
78
The Pumping Lemma for CFL
The pumping lemma for regular languages states that every sufficiently long string in a
regular language contains a short substring that can be pumped. That is, inserting as many
copies of the substring as we like always yields a string in the regular language.
The pumping lemma for CFL’s states that there are always two short substrings close
together that can be repeated, both the same number of times, as often as we like.
For example, consider a CFL L={a
n
b
n
 n > 1}. Equivalent CNF grammar is having
productions S ÷ AC  AB, A ÷ a, B ÷ b, C ÷ SB. The parse tree for the string a
4
b
4
is
given in figure 1
.
Both leftmost derivation and rightmost derivation have same parse tree
because the grammar is unambiguous.
Extend the tree by duplicating the terminals generated at each level on all lower levels. The
extended parse tree for the string a
4
b
4
is given in figure 2. Number of symbols at each level
is at most twice of previous level. 1 symbols at level 0, 2 symbols at 1, 4 symbols at 2 …2
i
symbols at level i. To have 2
n
symbols at bottom level, tree must be having at least depth of
n and level of at least n+1.
Pumping Lemma Theorem:
Figure 1: Parse tree for a
4
b
4
Figure 2: Extended Parse tree for a
4
b
4
79
Let L be a CFL. Then there exists a constant k> 0 such that if z is any string in L such that z
> k, then we can write z = uvwxy such that
1. vwx s k (that is, the middle portion is not too long).
2. vx = c (since v and x are the pieces to be “pumped”, at least one of the strings we
pump must not be empty).
3. For all i > 0, uv
i
wx
i
y is in L.
Proof:
The parse tree for a grammar G in CNF will be a binary tree. Let k = 2
n+1
, where n is the
number of variables of G. Suppose ze L(G) and z > k. Any parse tree for z must be of depth
at least n+1. The longest path in the parse tree is at least n+1, so this path must contain at
least n+1 occurrences of the variables. By pigeonhole principle, some variables occur more
than once along the path. Reading from bottom to top, consider the first pair of same variable
along the path. Say X has 2 occurrences. Break z into uvwxy such that w is the string of
terminals generated at the lower occurrence of X and vwx is the string generated by upper
occurrence of X.
Example parse tree:
For the above example S has repeated occurrences, and the parse tree is shown in figure 3. w = ab is the string generated by lower
occurrence of S and vwx = aabb is the string generated by upper occurrence of S. So here u=aa, v=a, w=ab, x=b, y=bb.
Figure 3: Parse tree for a
4
b
4
with
repeated occurrences of S circled.
Figure 4: sub trees
80
Let T be the subtree rooted at upper occurrence of S and t be subtree rooted at lower
occurrence of S. These parse trees are shown in figure 4. To get uv
2
wx
2
y eL, cut out t and
replace it with copy of T. The parse tree for uv
2
wx
2
y eL is given in figure 5. Cutting out t
and replacing it with copy of T as many times to get a valid parse tree for uv
i
wx
i
y for i > 1.
To get uwy e L, cut T out of the original tree and replace it with t to get a parse tree of
uv
0
wx
0
y = uwy as shown in figure 6.
Pumping Lemma game:
1. To show that a language L is not a CFL, assume L is context free.
2. Choose an “appropriate” string z in L
3. Express z = uvwxy following rules of pumping lemma
4. Show that uv
k
wx
k
y is not in L, for some k
5. The above contradicts the Pumping Lemma
6. Our assumption that L is context free is wrong
Example:
Show that L = {a
i
b
i
c
i
 i >1} is not CFL
Solution:
Assume L is CFL. Choose an appropriate z = a
n
b
n
c
n
= uvwxy. Since vwx s n then vwx can
either consists of
Figure 5: Parse tree for uv
2
wx
2
y
eL
Figure 6: Parse tree for uwy e L
81
1. All a’s or all b’s or all c’s
2. Some a’s and some b’s
3. Some b’s and some c’s
Case 1: vwx consists of all a’s
If z = a
2
b
2
c
2
and u = c, v = a, w = c, x = a and y = b
2
c
2
then, uv
2
wx
2
y will be a
4
b
2
c
2
=L
Case 2: vwx consists of some a’s and some b’s
If z = a
2
b
2
c
2
and u = a, v = a, w = c, x = b, y = bc
2
, then uv
2
wx
2
y will be a
3
b
3
c
2
=L
Case 3: vwx consists of some b’s and some c’s
If z = a
2
b
2
c
2
and u = a
2
b, v = b, w = c, x = c, y = c, then uv
2
wx
2
y will be a
2
b
3
c
2
=L
If you consider any of the above 3 cases, uv
2
wx
2
y will not be having an equal number of a’s,
b’s and c’s. But Pumping Lemma says uv
2
wx
2
y eL. Can’t contradict the pumping lemma!
Our original assumption must be wrong. So L is not contextfree.
Example:
Show that L = {ww w e{0, 1}*} is not CFL
Solution:
Assume L is CFL. It is sufficient to show that L1= {0
m
1
n
0
m
1
n
 m,n > 0}, where n is
pumping lemma constant, is a CFL. Pick any z = 0
n
1
n
0
n
1
n
= uvwxy, satisfying the
conditions vwx s n and vx =c.
This language we prove by taking the case of i = 0, in the pumping lemma satisfying the
condition uv
i
wx
i
y for i >0.
z is having a length of 4n. So if vwx s n, then uwy > 3n. According to pumping lemma,
uwy e L. Then uwy will be some string in the form of tt, where t is repeating. If so, n t >
3n/2.
Suppose vwx is within first n 0’s: let vx consists of k 0’s. Then uwy begins with 0
nk
1
n
uwy = 4nk. If uwy is some repeating string tt, then t =2nk/2. t does end in 0 but tt ends
with 1. So second t is not a repetition of first t.
Example: z = 0
3
1
3
0
3
1
3
,
vx = 0
2
then uwy = tt = 01
3
0
3
1
3
, so first t = 01
3
0 and second t =
0
2
1
3.
Both t’s are not same.
82
Suppose vwx consists of 1
st
block of 0’s and first block of 1’s: vx consists of only 0’s if
x= c, then uwy is not in the form tt. If vx has at least one 1, then t is at least 3n/2 and first t
ends with a 0, not a 1.
Very similar explanations could be given for the cases of vwx consists of first block of 1’s
and vwx consists of 1
st
block of 1’s and 2
nd
block of 0’s. In all cases uwy is expected to be in
the form of tt. But first t and second t are not the same string. So uwy is not in L and L is not
context free.
Example:
Show that L={0
i
1
j
2
i
3
j
 i > 1, j > 1} is not CFL
Solution:
Assume L is CFL. Pick z = uvwxy = 0
n
1
n
2
n
3
n
where vwx s n and vx = c. vwx can consist
of a substring of one of the symbols or straddles of two adjacent symbols.
Case 1: vwx consists of a substring of one of the symbols
Then uwy has n of 3 different symbols and fewer than n of 4
th
symbol. Then uwy is not in L.
Case 2: vwx consists of 2 adjacent symbols say 1 & 2
Then uwy is missing some 1’s or 2’s and uwy is not in L.
If we consider any combinations of above cases, we get uwy, which is not CFL. This
contradicts the assumption. So L is not a CFL.
Exercises:
Using pumping lemma for CFL prove that below languages are not context free
1. {0
p
 p is a prime}
2. {a
n
b
n
c
i
 i s n}
83
Closure Properties of CFL
Many operations on Context Free Languages (CFL) guarantee to produce CFL. A few do not
produce CFL. Closure properties consider operations on CFL that are guaranteed to produce
a CFL. The CFL’s are closed under substitution, union, concatenation, closure (star),
reversal, homomorphism and inverse homomorphism. CFL’s are not closed under
intersection (but the intersection of a CFL and a regular language is always a CFL),
complementation, and setdifference.
I. Substitution:
By substitution operation, each symbol in the strings of one language is replaced by an entire
CFL language.
Example:
S(0) = {a
n
b
n
 n >1}, S(1)={aa,bb} is a substitution on alphabet E ={0, 1}.
Theorem:
If a substitution s assigns a CFL to every symbol in the alphabet of a CFL L, then s(L) is a
CFL.
Proof:
Let G = (V, E, P, S) be grammar for the CFL L. Let G
a
= (V
a
, T
a
, P
a
, S
a
) be the grammar
corresponding to each terminal a e E and V · V
a
= . Then G'= (V', T', P', S) is a grammar
for s(L) where
 V' = V V
a
 T'= union of T
a
’s
all for a e E
 P' consists of
o All productions in any P
a
for a e E
o The productions of P, with each terminal a is replaced by S
a
everywhere a
occurs.
Example:
L = {0
n
1
n
 n > 1}, generated by the grammar S ÷ 0S1  01, s(0) = {a
n
b
m
 m s n}, generated
by the grammar S ÷ aSb  A; A ÷ aA  ab, s(1) = {ab, abc}, generated by the grammar S ÷
abA, A ÷ c c. Rename second and third S’s to S
0
and S
1
, respectively. Rename second A to
B. Resulting grammars are:
84
S ÷ 0S1  01
S
0
÷ aS
0
b  A; A ÷ aA  ab
S
1
÷ abB; B ÷ c  c
In the first grammar replace 0 by S
0
and 1 by S
1
. The resulted grammar after substitution is:
S ÷ S
0
SS
1
 S
0
S
1
S
0
÷ aS
0
b  A; A ÷aA  ab S
1
÷abB; B÷ c  c
II. Application of substitution:
a. Closure under union of CFL’s L
1
and L
2
:
Use L={a, b}, s(a)=L
1
and s(b)=L
2
. Then s(L)= L
1
L
2
.
How to get grammar for L
1
L
2
?
Add new start symbol S and rules S ÷ S
1
 S
2
The grammar for L
1
L
2
is
G = (V, T, P, S) where V = {V
1
V
2
S}, Se (V
1
V
2)
and P = {P
1
P
2
{S ÷ S
1
 S
2
}}
Example:
L
1
= {a
n
b
n
 n > 0}, L
2
= {b
n
a
n
 n > 0}. Their corresponding grammars are
G
1
: S
1
÷ aS
1
b  c, G
2
:
S
2
÷ bS
2
a  c
The grammar for L
1
L
2
is
G = ({S, S
1,
S
2
}, {a, b}, {S ÷ S
1
 S
2
, S
1
÷ aS
1
b  c, S
2
÷ bS
2
a}, S).
b. Closure under concatenation of CFL’s L
1
and L
2
:
Let L={ab}, s(a)=L
1
and s(b)=L
2
. Then s(L)=L
1
L
2
How to get grammar for L
1
L
2
?
Add new start symbol and rule S ÷ S
1
S
2
The grammar for L
1
L
2
is G = (V, T, P, S) where V = V
1
V
2
{S}, S e V
1
V
2
and P
= P
1
P
2
{S ÷ S
1
S
2
}
85
Example:
L
1
= {a
n
b
n
 n > 0}, L
2
= {b
n
a
n
 n > 0} then L
1
L
2
= {a
n
b
{n+m}
a
m
 n, m > 0}
Their corresponding grammars are
G
1
: S
1
÷ aS
1
b  c, G
2
:
S
2
÷ bS
2
a  c
The grammar for L
1
L
2
is
G = ({S, S
1,
S
2
}, {a, b}, {S ÷ S
1
S
2
, S
1
÷ aS
1
b  c, S
2
÷ bS
2
a}, S).
c. Closure under Kleene’s star (closure * and positive closure
+
) of CFL’s L
1
:
Let L = {a}* (or L = {a}
+
) and s(a) = L
1
. Then s(L) = L
1
* (or s(L) = L
1
+
).
Example:
L
1
= {a
n
b
n
 n > 0} (L
1
)* = {a
{n1}
b
{n1}
... a
{nk}
b
{nk}
 k > 0 and ni > 0 for all i}
L
2
= {a
{n
2
}
 n > 1}, (L
2
)* = a*
How to get grammar for (L
1
)*:
Add new start symbol S and rules S ÷ SS
1
 c.
The grammar for (L
1
)*
is
G = (V, T, P, S), where V = V
1
{S}, S e V
1,
P= P
1
{S ÷ SS
1
 c}
d. Closure under homomorphism of CFL L
i
for every a
i
e¿:
Suppose L is a CFL over alphabet ¿ and h is a homomorphism on ¿. Let s be a substitution
that replaces every a
e ¿, by h(a). ie s(a) = {h(a)}. Then h(L) = s(L). ie h(L) ={h(a
1
)…h(a
k
)
 k > 0} where h(a
i
) is a homomorphism for every a
i
e ¿.
III. Closure under Reversal:
L is a CFL, so L
R
is a CFL. It is enough to reverse each production of a CFL for L, i.e., to
substitute each production A÷o by A÷o
R
.
IV. Intersection:
The CFL’s are not closed under intersection
86
Example:
The language L = {0
n
1
n
2
n
 n > 1} is not contextfree. But L
1
= {0
n
1
n
2
i
 n > 1, i > 1} is a
CFL and L
2
= {0
i
1
n
2
n
 n > 1, i > 1} is also a CFL. But L = L1
·
L
2.
Corresponding grammars for L
1
: S÷AB; A÷0A1  01; B÷2B  2 and corresponding
grammars for L
2
: S ÷AB; A÷0A  0; B÷1B2  12.
However, L = L
1 ·
L
2
, thus intersection of CFL’s is not CFL
a. Intersection of CFL and Regular Language:
Theorem: If L is CFL and R is a regular language, then L
·
R is a CFL.
Proof:
P = (Q
P
, ¿, I, o
P
, q
P
, Z
0
, F
P
) be PDA to accept L by final state. Let A = (Q
A
, ¿, o
A
, q
A
, F
A
)
for DFA to accept the Regular Language R. To get L
·
R, we have to run a Finite Automata
in parallel with a push down automata as shown in figure 1. Construct PDA P' = (Q, ¿, I,
o, q
o
, Z
0
, F) where
 Q = (Q
p
X Q
A
)
 q
o
= (q
p
, q
A
)
 F = (F
P
X F
A
)
 o is in the form o ((q, p), a, X) = ((r, s), g) such that
1. s = o
A
(p, a)
2. (r, g) is in o
P
(q, a, X)
That is for each move of PDA P, we make the same move in PDA P' and also we carry along
the state of DFA A in a second component of P'. P' accepts a string w if and only if both P
and A accept w. ie w is in L
·
R. The moves ((q
p
, q
A
), w, Z) *P' ((q, p), c, ¸) are possible if
and only if (q
p
, w, Z) *P (q, c,¸) moves and p = o*(q
A
, w) transitions are possible.
b. CFL and RL properties:
Stack
PDA
FA
AND
Accept/
Reject
Stack
PDA
FA
AND
Accept/
Reject
PDA
FA
AND
PDA
FA
AND
Accept/
Reject
Figure 1: PDA for L
·
R
87
Theorem: The following are true about CFL’s L, L
1
, and L
2
, and a regular language R.
1. Closure of CFL’s under setdifference with a regular language. ie L  R is a CFL.
Proof:
R is regular and regular language is closed under complement. So R
C
is also regular. We
know that L  R = L · R
C
. We have already proved the closure of intersection of a CFL
and a regular language. So CFL is closed under set difference with a Regular language.
2. CFL is not closed under complementation
L
C
is not necessarily a CFL
Proof:
Assume that CFLs were closed under complement. ie if L is a CFL then L
C
is a CFL.
Since CFLs are closed under union, L
1
C
L
2
C
is a CFL. By our assumption (L
1
C
L
2
C
)
C
is a CFL. But (L
1
C
L
2
C
)
C
= L
1
· L
2,
which we just showed isn’t necessarily
a CFL. Contradiction! . So our assumption is false. CFL is not closed under
complementation.
3. CFLs are not closed under setdifference. ie L
1
 L
2
is not necessarily a CFL.
Proof:
Let L1 = ¿*  L. ¿* is regular and is also CFL. But ¿*  L = L
C
. If CFLs were closed
under set difference, then ¿*  L = L
C
would always be a CFL. But CFL’s are not
closed under complementation. So CFLs are not closed under setdifference.
V. Inverse Homomorphism:
Recall that if h is a homomorphism, and L is any language, then h
1
(L), called an inverse
homomorphism, is the set of all strings w such that h(w) e L. The CFL’s are closed under
inverse homomorphism.
Theorem:
If L is a CFL and h is a homomorphism, then h
1
(L) is a CFL
88
89
We can prove closure of CFL under inverse homomorphism by designing a new PDA as
shown in figure 2. After input a is read, h(a) is placed in a buffer. Symbols of h(a) are used
one at a time and fed to PDA being simulated. Only when the buffer is empty does the PDA
read another of its input symbol and apply homomorphism to it.
Suppose h applies to symbols of alphabet Σ and produces strings in T*. Let PDA P = (Q, T,
Γ, δ, q
0
, Z
0
, F) that accept CFL L by final state. We construct a new PDA P' = (Q', Σ, Γ, δ',
(q
0
, c), Z
0
, (F x c)) to accept h
1
(L), where
 Q' is the set of pairs (q, x) such that
o q is a state in Q
o x is a suffix of some string h(a) for some input string a in Σ
 δ' is defined by
o δ' ((q, c), a, X) = {((q, h(a)),X)}
o If δ(q, b, X) = {(p, ¸)} where be T or b = c then δ' ((q, bx), c, X) = {((p,
x), ¸)}
 The start state of P' is (q
0
, c)
 The accepting state of P' is (q, c), where q is an accepting state of P.
Once we accept the relationship between P and P', P accepts h(w) if and only if P' accepts w,
because of the way the accepting states of P' are defined.
Thus L(P')=h
1
(L(P))
Stack
PDA
h
Accept/
Reject
Input
a
h(a)
Buffer
Stack
PDA
h
Accept/
Reject
Input
a
h(a)
Buffer
Figure 2: PDA to simulate inverse homomorphism
90
Language Hierarchy
and
History of Turing Machine
A Hierarchy of Formal Languages
Regular language
A language is called a regular language if some finite automaton recognizes it.
Ex: To recognize string that is a multiple of 4
Turing
machines
PDA
DFA
r.e. language
Context free
language
\
Regular language
0
S2
S1
0
S0
0
1
0
1
91
But can regular language recognize strings of the form 0
n
1
n
? No
Context Free Language
A language is called Context free language iff some pushdown automaton recognizes it.
Ex: To recognize a string of the form 0
n
1
n
Limitation again
Can context free language recognize strings of the form 0
n
1
n
2
n
? No
Recursively Enumerable Language (r.e. language)
A language is called r.e language if some Turing machine recognizes it.
Ex: To recognize strings of the form 0
n
1
n
2
n
S
A
: Accepting state
ì / 1 0S S ÷
) (
) (
) 0 0 (
) (
) 1 1 (
) ) 2 (
) 1 1 (
) 1 (
) 0 0 (
) 0 (
0
0 3
3 3
3 3
3 3
3 2
2 2
2 1
1 1
1 0
,Y,R (S ,Y) S
,X,R (S ,X) S
,L , (S ) , S
,Y,L (S ,Y) S
,L , (S ) , S
,Z,L (S , S
,R , (S ) , S
,Y,R (S ) , S
,R , (S ) , S
,X,R (S ) , S
A ÷
÷
÷
÷
÷
÷
÷
÷
÷
÷
92
Formal Machines Overview
DFA
NFA
NFA with ìmoves
2way DFA
DFA
Regular Lang.
Regular Expn.
Regex
operators:
* (Kleene *)
 (choice)
. (concat)
Regex
example:
(0  1)*
93
But there are strings that cannot be recognized by PDAs. For example: a
n
b
n
c
n
Some Historical Notes
At the turn of the 20
th
century the German mathematician David Hilbert proposed the
Entscheidungsproblem. Given a set X, and a universe of elements U, and a criteria for
membership, is it possible to formulate a general procedureto decide whether a given
element of U is a member of X?
Hilbert’s grand ideas were watered down by the incompleteness theorem proposed by
the Austrian Kurt Gödel. His theorem states that in any mathematical system, there exist
certain obviously true assertions that cannot be proved to be true by the system.
In 1936 the British cryptologist Alan Turing, addressed Hilbert’s
Entscheidungsproblem using a different approach. He proposed two kinds of mathematical
PDA
CFL
CFG
CFL
example:
0
n
1
n
PDA
Multistack PDA PDA with queues
94
machines called the amachine and the cmachine respectively, and showed the existence of
some problems where membership cannot be determined.
But Turing’s amachine became famous for something else other than its original
intentions. The amachine was found to be more expressive than CFGs. It can recognize
strings that cannot be modeled by CFLs like a
n
b
n
c
n
. The amachines came to be more
popularly known as Turing Machines
The Princeton mathematician Alonzo Church recognized the power of amachines.
He invited Turing to Princeton to compare Turing Machines with his own lcalculus.
Church and Turing proved the equivalence of Turing Machines and lcalculus, and
showed that they represent algorithmic computation. This is called the ChurchTuring thesis.
Turing Machines
Definition:
A Turing Machine (TM) is an abstract, mathematical model that describes what can
and cannot be computed. A Turing Machine consists of a tape of infinite length, on which
input is provided as a finite sequence of symbols. A head reads the input tape. The Turing
Machine starts at “start state” S
0
. On reading an input symbol it optionally replaces it with
another symbol, changes its internal state and moves one cell to the right or left.
Notation for the Turing Machine :
TM = <S, T, S
0
, o, H> where,
S is a set of TM states
T is a set of tape symbols
S
0
is the start state
H c S is a set of halting states
o : S x T S x T x {L,R}
is the transition function
{L,R} is direction in which the
head moves
L : Left R: Right
95
input symbols on infinite length tape
head
The Turing machine model uses an infinite tape as its unlimited memory. (This is
important because it helps to show that there are tasks that these machines cannot perform,
even though unlimited memory and unlimited time is given.) The input symbols occupy
some of the tape’s cells, and other cells contain blank symbols.
Some of the characteristics of a Turing machine are:
1. The symbols can be both read from the tape and written on it.
2. The TM head can move in either directions – Left or Right.
3. The tape is of infinite length
4. The special states, Halting states and Accepting states, take immediate effect.
Solved examples:
TM Example 1:
Turing Machine U+1:
Given a string of 1s on a tape (followed by an infinite number of 0s), add one more 1
at the end of the string.
Input : #111100000000…….
Output : #1111100000000……….
Initially the TM is in Start state S
0
. Move right as long as the input symbol is 1. When a 0 is
encountered, replace it with 1 and halt.
Transitions:
(S
0
, 1) (S
0
, 1, R)
(S
0
, 0) ( h , 1, STOP)
TM Example 2 :
1 0 1 0 1 1 1 1 1 1 0
96
TM: XY
Given two unary numbers x and y, compute xy using a TM. For purposes of
simplicity we shall be using multiple tape symbols.
Ex: 5 (11111) – 3 (111) = 2 (11)
#11111b1110000…..
#___11b___000…
a) Stamp out the first 1 of x and seek the first 1 of y.
(S
0
, 1) (S
1
, _, R)
(S
0
, b) (h, b, STOP)
(S
1
, 1) (S
1
, 1, R)
(S
1
, b) (S
2
, b, R)
b) Once the first 1 of y is reached, stamp it out. If instead the input ends, then y has finished.
But in x, we have stamped out one extra 1, which we should replace. So, go to some state s5
which can handle this.
(S
2
, 1) (S
3
, _, L)
(S
2
,_) (S
2
, _, R)
(S
2
, 0) (S
5
, 0, L)
c) State s3 is when corresponding 1s from both x and y have been stamped out. Now go back
to x to find the next 1 to stamp. While searching for the next 1 from x, if we reach the head of
tape, then stop.
(S
3
, _) (S
3
, _, L)
(S
3
,b) (S
4
, b, L)
(S
4
, 1) (S
4
, 1, L)
(S
4
, _) (S
0
, _, R)
(S
4
, #) (h, #, STOP)
d) State s5 is when y ended while we were looking for a 1 to stamp. This means we have
stamped out one extra 1 in x. So, go back to x, and replace the blank character with 1 and
stop the process.
(S
5
, _) (S
5
, _, L)
(S
5
,b) (S
6
, b, L)
(S
6
, 1) (S
6
, 1, L)
(S
6
, _) (h, 1, STOP)
97
Design of Turing Machines
and
Universal Turing Machine
Solved examples:
TM Example 1: Design a Turing Machine to recognize 0
n
1
n
2
n
ex: #000111222_ _ _ _ _…….
Step 1: Stamp the first 0 with X, then seek the first 1 and stamp it with Y, and then seek the
first 2 and stamp it with Z and then move left.
S
0
= Start State, seeking 0, stamp it with X
S1 = Seeking 1, stamp it with Y
S2 = Seeking 2, stamp it with Z
Step 2: Move left until an X is reached, then move one step right.
S3 = Seeking X, to repeat the process.
Step 3: Move right until the end of the input denoted by blank( _ ) is reached passing through
X Y Z s only, then the accepting state S
A
is reached.
) ) 2 (
) 1 1 (
) 1 (
) 0 0 (
) 0 (
3 2
2 2
2 1
1 1
1 0
,Z,L (S , S
,R , (S ) , S
,Y,R (S ) , S
,R , (S ) , S
,X,R (S ) , S
÷
÷
÷
÷
÷
) (
) 0 0 (
) (
) 1 1 (
0 3
3 3
3 3
3 3
,X,R (S ,X) S
,L , (S ) , S
,Y,L (S ,Y) S
,L , (S ) , S
÷
÷
÷
÷
) _ _ (
) (
) (
) (
4
4 4
4 4
4 0
,STOP , (S ) , S
,Z,R (S ,Z) S
,Y,R (S ,Y) S
,Y,R (S ,Y) S
A ÷
÷
÷
÷
98
S4 = Seeking blank
These are the transitions that result in halting states.
TM Example 2 : Design a Turing machine to accept a Palindrome
ex: #1011101_ _ _ _ _…….
Step 1: Stamp the first character (0/1) with _, then seek the last character by moving till a _ is
reached. If the last character is not 0/1 (as required) then halt the process immediately.
Step 2: If the last character is 0/1 accordingly, then move left until a blank is reached to start
the process again.
) _ _ (
) 2 2 (
) 2 2 (
) 1 1 (
) _ _ (
) 2 2 (
) 1 1 (
2
1
0
0
4
4
4
,STOP (h, ) , S
,STOP (h, ) , S
,STOP (h, ) , S
,STOP (h, ) , S
,STOP , (S ) , S
,STOP (h, ) , S
,STOP (h, ) , S
A
÷
÷
÷
÷
÷
÷
÷
) 0 ( 0 (
) _ _ (
) 1 ( 1 (
) _ _ (
) _ 1 (
) _ 0 (
5
5 2
3
3 1
2 0
1 0
,STOP h, ) , S
,L , (S ) , S
,STOP h, ) , S
,L , (S ) , S
,R , (S ) , S
,R , (S ) , S
÷
÷
÷
÷
÷
÷
) _ _) (
) 0 ) 0 (
) 1 ) 1 (
) _ ) 1 (
) _ _) (
) 0 ) 0 (
) 1 ) 1 (
) _ ) 0 (
0 6
6 6
6 6
6 5
0 4
4 4
4 4
4 3
,R , (S , S
,L , (S , S
,L , (S , S
,L , (S , S
,R , (S , S
,L , (S , S
,L , (S , S
,L , (S , S
÷
÷
÷
÷
÷
÷
÷
÷
99
Step 3 : If a blank ( _ ) is reached when seeking next pair of characters to match or when
seeking a matching character, then accepting state is reached.
The sequence of events for the above given input are as follows:
#s
0
10101_ _ _
#_s
2
0101_ _ _
#_0s
2
101_ _ _
. . . .
#_0101s
5
_ _ _
#_010s
6
_ _ _ _
#_s
6
0101_ _ _
#_s
0
0101_ _ _
. . . .
#_ _ _ _ s
5
_ _ _ _ _ _
#_ _ _ _ s
A
_ _ _ _ _ _
Exercises:
1. Design a TM to recognize a string of the form a
n
b
2n
.
2. Design a TM to recognize a string of 0s and 1s such that the number of 0s is not twice as
that of 1s.
Modularization of TMs
Designing complex TM s can be done using modular approach. The main problem
can be divided into sequence of modules. Inside each module, there could be several state
transitions.
For example, the problem of designing Turing machine to recognize the language 0
n
1
n
2
n
can
be divided into modules such as 0stamper, 1stamper, 0seeker, 1seeker, 2seeker and 2
stamper. The associations between the modules are shown in the following figure:
) _ _) (
) _ _) (
) _ _) (
0
5
3
,STOP , (S , S
,STOP , (S , S
,STOP , (S , S
A
A
A
÷
÷
÷
100
Universal Turing Machine
A Universal Turing Machine UTM takes an encoding of a TM and the input data as its input
in its tape and behaves as that TM on the input data.
A TM spec could be as follows:
TM = (S,S0,H,T,d)
Suppose, S={a,b,c,d}, S0=a, H={b,d} T={0,1}
o : (a,0) (b,1,R) , (a,1) (c,1,R) ,
(c,0) (d,0,R) and so on
then TM spec:
$abcd$a$bd$01$a0b1Ra1c1Rc0d0R…….
where $ is delimiter
This spec along with the actual input data would be the input to the UTM.
This can be encoded in binary by assigning numbers to each of the characters appearing in
the TM spec.
TM: 0
n
1
n
2
n
0Stamper 1Seeker
0Seeker
1Stamper
2Seeker
2Stamper
101
The encoding can be as follows:
$ : 0000 0 : 0101
a : 0001 1 : 0110
b : 0010 L : 0111
c : 0011 R : 1000
d : 0100
So the TM spec given in previous slide can be encoded as:
0000.0001.0010.0011.0100.0000.0001.0000.0010.0100 ……
Hence TM spec can be regarded just as a number.
Sequence of actions in UTM:
Initially UTM is in the start state S0.
Load the input which is TM spec.
Go back and find which transition to apply.
Make changes, where necessary.
Then store the changes.
Then repeat the steps with next input.
Hence, the sequence goes through the cycle:
Store Execute Decode Load ÷ ÷ ÷
102
Extensions to Turing Machines
Proving Equivalence
For any two machines M
1
from class C
1
and M
2
from class C
2
:
M
2
is said to be at least as expressive as M
1
if L(M
2
) = L(M
1
) or if M
2
can simulate M
1
.
M
1
is said to be at least as expressive as M
2
if L(M
1
) = L(M
2
) or if M
1
can simulate M
2
.
Composite Tape TMs
Track 0
Track 1
A composite tape consists of many tracks which can be read or written simultaneously.
A composite tape TM (CTM) contains more than one tracks in its tape.
Equivalence of CTMs and TMs
A CTM is simply a TM with a complex alphabet..
T = {a, b, c, d}
T’ = {00, 01, 10, 11}
Turing Machines with Stay Option
Turing Machines with stay option has a third option for movement of the TM head:
left, right or stay.
STM = <S, T, o, s
0
, H>
o: S x T à S x T x {L, R, S}
Equivalence of STMs and TMs
STM = TM:
Just don’t use the S option…
TM = STM:
For L and R moves of a given STM build a TM that moves correspondingly L or R…
TM = STM:
For S moves of the STM, do the following:
1.Move right,
2.Move back left without changing the tape
0 1 1 0 1 0 1 0 0 …
0 0 1 1 1 1 1 1 0 …
103
3.STM: o(s,a)  (s’,b,S)
TM: o(s,a)  (s’’, b, R)
o(s’’,*)  (s’,*,L)
2way Infinite Turing Machine
In a 2way infinite TM (2TM), the tape is infinite on both sides.
There is no # that delimits the left end of the tape.
Equivalence of 2TMs and TMs
2TM = TM:
Just don’t use the left part of the tape…
TM = 2TM:
Simulate a 2way infinite tape on a oneway infinite tape…
Multitape Turing Machines
A multitape TM (MTM) utilizes many tapes.
… 6 5 4 3 2 1 0 1 2 3 4 5 6 …
0 –1 1 –2 2 –3 3 –4 4 –5 5 …
104
Equivalence of MTMs and TMs
MTM = TM:
Use just the first tape…
TM = MTM:
Reduction of multiple tapes to a single tape.
Consider an MTM having m tapes. A single tape TM that is equivalent can be constructed by
reducing m tapes to a single tape.
Multidimensional TMs
Multidimensional TMs (MDTMs) use a multidimensional space instead of a single
dimensional tape.
0 1 2 3 4 5 6 7 …
0 1 2 3 4 5 6 7 …
0 1 2 3 4 5 6 7 …
A0 B0 C0 A1 B1 C1 A2 B2 C2 A3 B3 ..
A
B
C
TM
TM
105
Equivalence of MDTMs to TMs
Reducing a multidimensional space to a single dimensional tape.
0 1 2 3 4 5 6 7 8 9 10
0
1
2
3
4
Nondeterministic TM
A nondeterministic TM (NTM) is defined as:
NTM = <S, T, s
0
, o, H>
where o: S x T à2
SxTx{L,R}
Ex: (s
2
,a) à {(s
3
,b,L) (s
4
,a,R)}
Equivalence of NTMs and TMs
A “concurrent” view of an NTM:
(s
2
,a) à {(s
3
,b,L) (s
4
,a,R)}
è at (s
2
,a), two TMs are spawned:
(s
2
,a) à (s
3
,b,L)
(s
2
,a) à (s
4
,a,R)
106
Simulating an NTM with an MDTM
Consider an MDTM to simulate an NTM:
(s
2
,a) à {(s
3
,b,L) (s
4
,a,R)}
bccaaabccacb
s
2
bccbaabccacb
s
3
bccaaabccacb
s
4
Draw a DFA to accept string of 0’s and 1’s ending with the string 011.
1 q0
0
0 q1
1 0
q2 0
1
q3
Obtain a DFA to accept strings of a’s and b’s having a sub string aa
1
b q0 a b q1 a q2
a,b
Obtain a DFA to accept strings of a’s and b’s except those containing the substring aab.
b q0 a b q1 a q2
a b q3
a,b
Obtain DFAs to accept strings of a’s and b’s having exactly one a, b q0 a b q0 b q0 a b q1 a b q2 a a q1 b q3 a q4 a, b b q1 a q2 a, b a,b
Obtain a DFA to accept strings of a’s and b’s having even number of a’s and b’s
The machine to accept even number of a’s and b’s is shown in fig.2.22. a q
0
q a b a b q
a
3 1
b q
2
b
2
Fig.2.22 DFA to accept even no. of a’s and b’s a q0 a b q2 a a q0 a b q2 a a q0 a b q2 a Regular language Definition: Let M = (Q, , , q0, A) be a DFA. The language L is regular if there exists a machine M such that L = L(M). * Applications of Finite Automata * String matching/processing Compiler Construction The various compilers such as C/C++, Pascal, Fortran or any other compiler is designed using the finite automata. The DFAs are extensively used in the building the various phases of compiler such as Lexical analysis (To identify the tokens, identifiers, to strip of the comments etc.) Syntax analysis (To check the syntax of each statement or control statement used in the program) Code optimization (To remove the un wanted code) 3 b a q3 b b q1 b a q3 b b q1 b a q3 b b q1
is transition function from Q x {U} to 2Q. The string (also called language) w accepted by an NFA can be defined in formal notation as: L(M) = { w  w *and *(q0. This section lists some applications: 1. remote sensing or controller etc. Large natural vocabularies can be described using finite automaton which includes the applications such as spelling checkers and advisers. Non deterministic finite automata(NFA) Definition: An NFA is a 5tuple or quintuple M = (Q. multilanguage dictionaries. finite set of input alphabets. 4. A Q is set of final states. automatic traffic signals. . . A) be a DFA where Q is set of finite states. in calculators to evaluate complex expressions based on the priority of an operator etc. 3. 2. computer graphics. Finite automaton is very useful in hardware design such as circuit verification. In game theory and games wherein we use some control characters to fight against a monster. finite automaton plays a very important role. This function shows the change of state from one state to a set of states based on the input symbol. . automatic sensors. to indent the documents. to name a few. economics. we can find the approximate solutions. q0 Q is the start state. Finite automaton is very useful in recognizing difficult problems i. w) = Q with atleast one 4 . linguistics etc. in design of the hardware board (mother board or any other hardware unit). radio controlled toys. q0. Code generation (To generate the machine code) Other applications The concept of finite automata is used in wide applications. q0 is the start state and A is the final or accepting state. using theory of computation. Even though there is no general solution exists for the specified problem. q0. Any editor that we use uses finite automaton for implementation. It is not possible to list all the applications as there are infinite number of applications... elevators. is non empty. is transition function which is a mapping from Q x { U } to subsets of 2Q. A) where Q is non empty. finite set of states. sometimes it is very essential to solve an undecidable problem. .e. is set of input alphabets (from which a string can be formed). Acceptance of language Definition: Let M = (Q.
qc AN i. qj. AN) be an NFA and accepts the language L(MN). N.…. The procedure to convert an NFA to its equivalent DFA is shown below: Step1: The start state of NFA MN is the start state of DFA MD. a) U ……N(qk. 3.….qk] to [ql.qk].qn] is added to QD in the previous step.qn] on the input symbol a iff the state [ql. at least one of the component in [qa.….….…. So.qn] say. D.qc] should be the final state of NFA. the transitions for each input symbol in can be obtained as shown below: 1. Step4: If epsilon () is accepted by NFA. Step2: For each state [qi. if it is not already in QD. qb.qn] to QD. qm. qb. a) = N(qi. There should be an equivalent DFA MD = (QD..1 Conversion from NFA to DFA Let MN = (QN.qk] in QD.…. qm. q0. Step3: The state [qa.e. qj. The way to obtain different transitions is shown in step2. qm. qm. then start state q0 of DFA is made the final state. ….Component of Q in A} Obtain an NFA to accept the following language L = {w  w ababn or aban where n 0} The machine to accept either ababn or aban where n 0 is shown below: b q0 q1 a q2 b q3 a q4 a q5 a q6 b q7 2. AD) such that L(MD) = L(MN).…. D([qi.qc] QD is the final state. a) U N(qj. 2. add q0(which is the start state of NFA) to QD and find the transitions from this state. D. if at least one of the state in qa.…. a) = [ql. Add the state [ql. N.…. Add the transition from [qi. qb. 5 . q0. Convert the following NFA into an equivalent DFA. qj..
1 q 1 2 1 Step1: q0 is the start of DFA (see step1 in the conversion procedure). q1] [q1] Consider the state [q0. 0) U N(q1. q1} U {q2} = [q0.11) When a = 1 D D([q0. 1) F A D F A 0 [q0. q1]: When a = 0 D([q0. So. q1]. Consider the state [q0]: When a = 0 D([q0]. q2] (2. q1] 1 [q1] 6 . 0) = N(q0. q1] (2.8) = N([q0]. [q0] Q [q0.7) Step2: Find the new states from each state in QD and obtain the corresponding transitions. add these two states to QD so that QD = {[q0].7). [q0. 0) = N([q0].8) and (2.9) When a = 1 D([q0].1 q 0. = N([q0.10) The corresponding transitions on a = 0 and a = 1 are shown below. 0) = [q0. 0) 0) = {q0.9) are not in QD(2. 1) Since the states obtained in (2. q1. q1]. q1]. 1) = [q1] (2.0 q0 0. q1]. [q1] } (2. QD = {[q0]} (2. = N([q0. q1].
q2] } and add the transitions on a = 0 and a = 1 as shown below: Q [q0] [q0. q1]. [q1. q2] 0 [q0. q2].15) When a = 1 D D([q1]. q1.15) are same and the state q2 is not in QD(see 2. [q0. q1]. q2] [q1.12) are the not defined in QD(see 2. 1) = [q2] (2.16) and add the transitions on a = 0 and a = 1 as shown below: [q0] [q0. [q0. [q2]} (2. q2] (2. q1. add the state q2 to QD so that QD = {[q0]. q1. q1. 1) = {q1} U {q2} = [q1.11) and (2. q2] 1 [q1] [q1. q1] 0 1 [q0. [q1. [q1]. q2] D F A 7 D F . q1] [q1] [q0. 1) U N(q1. [q0. q1.10). q1] [q1] [q0. 0) = [q2] D (2. [q0.13). q1] [q0.12) Since the states obtained in (2. q2]. 0) = N([q1].13) Consider the state [q1]: When a = 0 D([q1]. 1) F A Since the states obtained in (2.14) and (2. add these two states to QD so that QD = {[q0]. q2] (2.1) = N(q0. [q1]. q2] [q1. q2].14) F A = N([q1].
q2]. 0) U N(q2.q2]. 1) U N(q1.q1. 0) = N([q1. D 1) F A = = = = N([q0.q2]: When a = 0 D([q1.q2] (2.q1. 0) U N(q1.q1. 0) 8 D F A . q2] [q2] [q0.q2]. q2] Q D F A Consider the state [q1.q1. q2] [q2] [q2] Consider the state [q0. 1) U N(q2.Q [q1] [q2] [q0. see 2. q1. q1] [q0.16). 1) N(q0. q2] [q2] 0 [q0.q2]. But.18) are not new states (are already in QD. q2] [q1.17) and (2. q1] [q1] [q0.q1.q1} U {q2} U {} [q0. 0) N(q0. 0) {q0. 0) = = = = N([q0.17) When a = 1 D([q0.18) Since the states obtained in (2.q2] 1 [q1] [q1. q2] [q1.q1. q2] (2.q2]. the transitions on a = 0 and a = 1 should be added to the transitional table as shown below: [q0] [q0. q2] [q2] [q1. 1) {q1} U {q2} U {q2} [q1. q1. do not add these two states to QD.q1.q2]: When a = 0 D([q0.q2]. q1.
21) D F = N([q2].q2] [q2] 1 [q1] [q1.35. 1) N(q1. q2] [q1.20) Since the states obtained in (2. 1) = = = = N([q1.22) When a = 1 D([q2].= N(q1. q2] [q2] Q Consider the state [q2]: When a = 0 D([q2].19) and (2.20) are not new states (are already in QD see 2. 0) = {q2} U {} = [q2] (2.16). q1. 0) = N([q2]. 1) U N(q2. A do not add these two states to QD. q1. A 1) = [q2] (2. see 2.16). 1) D F Since the states obtained in (2. 9 D F .q1. But. 1) {q2} U {q2} [q2] (2. 0) U N(q2.q2]. the transitions on a = 0 and a = 1 should be added to the transitional table as shown below: [q0] [q0. q2] [q2] [q0.14. q1] [q0. q2] [q2] 0 [q0. and final DFA is shown in figure 2. 0) = {} (2.q2].22) are not new states (are already in QD. q1] [q1] [q0.21) and (2. do not add these two states to QD. But. The final transitional table is shown in table 2. q2] [q2] [q1.19) When a = 1 D([q1. the transitions on a = 0 and a = 1 should be added to the transitional table.
2.35 The DFA Convert the following NFA to its equivalent DFA. q2] [q2] 1 [q1] [q1.q2] [q1. q1] [q0. q1] [q1] 0 1 0.q1] [q1] 0 [q0. 0 a 1 b 2 3 6 b 7 4 a 5 8 9 Let QD = {0} Consider the state [A]: When input is a: (A) 10 . [q0] [q0. q 1.q2] [q2] [q2] 0 1 [q0. q 2] 1 [q1. 1 [q0.q1. q1. q2] [q2] [q2] [q0] [q0q1.q2] [q0. 1 [q2] 0 1 Fig. q2] [q2] [q1. q2] 0.
4. 8. 9} order) (D) (ascending This is because. 5. 8. due to transitions (or without giving any input) there can be transition to states 3. 4. a) = {} = N(1. a) = = = = N({2. 6} {3. in state 2. 3. 9.4. a) {5} {5.3. 6. b) Consider the state [B]: When input is a: (B.9}. b) = {} When input is b: ( A.4. 6.3. in state 5 due to transitions. 4. 9. 8. the states reachable are {8. b) (C) This is because. 4.(A. 6. 9} = D When input is b: ( C. b) N({2. (C.6.9 also. 4. b) {7} {7. a) = {1} (B) = N(0.9} When input is b: ( B. 6} {3. a) = N(0. 3. 4. 3.6. all these states are reachable from state 2. Therefore. 5.6.4.6.9} = C Consider the state [C]: When input is a: (C.4. 4. 9}(ascending order) 11 = = = = . 8. (B. 8. b) = {2} = {2.3. 7. a) = {3. Therefore. 3. 6}. 9. 6. b) = {2. So. 9}. a) = N(1.
6. a) {5} {5. 4.4.15. 4. 8.4. 9. b) = {3. 3. 8. 6. 4. 8. from state 7 the states that are reachable without any input (i.6. Therefore. 9} order) (D) (ascending When input is b: (D. 5. a) {5} {5. 4.6. 12 . we can stop at this point and the transition table for the DFA is shown in table 2.6.9}. 4.8. transition) are {8. 4. b) {7} {7. a) = = = = N({3. (C. 8.(E) This is because.4. 3. 8. b) = = = = N({3. a) = = = = N({3. b) {7} {7.8. 9. 4.5.7. 6.9}.5. 6. 7. 7. 7.9}. 9. 3. 9.4. b) = = = = N({3. 8. 3.e. 6. 6} {3. 9}(ascending order) (E) Since there are no new states. 6} {3. 9} = E Consider the state [D]: When input is a: (D. 9} order) (E) (ascending Consider the state [E]: When input is a: (E. 4. 6} {3.8.9}. 8.7. 8. 5. 3.8. 4. 9. 9}(ascending order) (D) When input is b: (E. 6. 6} {3. 6}.. 8. 4.
36 D a F AA a B b C a D a b b E b Fig. 3. A 1. The final transition diagram of DFA is shown in figure 2. 13 . is a regular expression denoting an empty language. then a.15 Transitional table F A The states C. R+S is a regular expression corresponding to the language LRULS. 4. D and E. 2.D and E are final states. since 9 (final state of NFA) is present in C.36 The DFA Regular Languages Regular expression D F Definition: A regular expression is recursively defined as follows. D (epsilon) is a regular expression indicates the language containing an empty string. a is aF regular expression which indicates the language containing only {a} If R A a regular expression denoting the language LR and S is a regular expression is denoting the language LS. 2. A B C D E a B D D D b C E E E Q D Table 2.
ba and bb. ab. c.1 Meaning of regular expressions Obtain a regular expression to accept a language consisting of strings of a’s and b’s of even length. Regular expressions (a+b)* Meaning Set of strings of a’s and b’s of any length including the NULL string. + + + abc Set of string consisting of at least one ‘a’ followed by string consisting of at least one ‘b’ followed by string consisting of at least one ‘c’. String of a’s and b’s of even length can be obtained by the combination of the strings aa.S is a regular expression corresponding to the language LR.1 shows some examples of regular expressions and the language corresponding to these regular expressions. The language may even consist of an empty string denoted by . (a+b)*aa(a+b) Set of strings of a’s and b’s having a sub string * aa. The table 3. (a+b)*abb Set of strings of a’s and b’s ending with the string abb ab(a+b)* Set of strings of a’s and b’s starting with the string ab.LS. R* is a regular expression corresponding to the language LR*. So. 5. aa*bb*cc* Set of string consisting of at least one ‘a’ followed by string consisting of at least one ‘b’ followed by string consisting of at least one ‘c’. R. The expressions obtained by applying any of the rules from 14 are regular expressions. the regular expression can be of the form (aa + ab + ba + bb)* 14 . (a+b)* (a + Set of strings of a’s and b’s ending with either a bb) or bb (aa)*(bb)*b Set of strings consisting of even number of a’s followed by odd number of b’s (0+1)*000 Set of strings of 0’s and 1’s ending with three consecutive zeros(or ending with 000) (11)* Set consisting of even number of 1’s Table 3.. a*b*c* Set of string consisting of any number of a’s(may be empty string also) followed by any number of b’s(may include empty string) followed by any number of c’s(may include empty string).b.
Then there exists a finite automaton M = (Q. .1. the corresponding machines to recognize these expressions are shown in figure 3. So.S and R* are regular expressions which clearly uses three operators ‘+’. a variety of regular expressions can be obtained for a language and all are equivalent. .2 Schematic representation of FA accepting L(R) In the definition of a regular expression it is clear that if R and S are regular expression. So. q0. ab. q0 (a) qf q0 (b) qf q0 a (c) qf Fig 3. ab.a.1. the regular expression can also be represented as (a+b) (aa + ab + ba + bb)* Note: Even though these two expression are seems to be different.1. where q is the start state and f is the final state of machine M. Proof: By definition. and a The schematic representation of a regular expression R to accept the language L(R) is shown in figure 3. So. . 3.2. then R+S and R. So. and a are regular expressions. String of a’s and b’s of odd length can be obtained by the combination of the strings aa. Obtain NFA from the regular expression Theorem: Let R be a regular expression.1 NFAs to accept . ba and bb preceded by either a or b. ba and bb followed by either a or b.The * closure includes the empty string. the regular expression can be of the form (aa + ab + ba + bb)* (a+b) String of a’s and b’s of odd length can also be obtained by the combination of the strings aa. Note: This regular expression can also be represented using set notation as L(R) = {(aa + ab + ba + bb)n  n 0} Obtain a regular expression to accept a language consisting of strings of a’s and b’s of odd length. L(R) q M f Fig 3. the language corresponding to those two expression is same. A) which accepts L(R). ‘‘ and 15 .c respectively.b and 3.
Case 1: R = R1 + R2. 1.’. Here. upon accepting L(R2). We can construct an NFA which accepts either L(R1)*) as shown in figure 3.4To accept the language L(R1 . q2. Since there is a transition.4 that the machine after accepting L(R1) moves from state q1 to f1. q0 q1 f1 M1 L(R1) 16 qf .b.3. Case 2: R = R1 . 3. the machine moves to f2 which is the final state. Case 3: R = (R1)*. L(R1) q0 q1 f1 q2 f2 M1 M2 L(R2) qf Fig. We can construct an NFA which accepts L(R1) followed by L(R2) which can be represented as L(R1 . 1. becomes the final state of machine M and accepts the language L(R1. 3.5. R2) It is clear from figure 3.‘. q1.a. R2.3 that the machine can either accept L(R1) or L(R2). 2.5. Thus.R2). It can also be represented as shown in figure 3. R2) as shown in figure 3. Let us take each case separately and construct equivalent machine. 2. q0 is the start state of the combined machine and qf is the final state of combined machine M.4. without any input there will be a transition from state f1 to state q2. f1) be a machine which accepts the language L(R1) corresponding to the regular expression R1. L(R1) L(R2) q1 M1 q2 M2 f1 f2 Fig. In state q2. Let M2 = (Q2.3 To accept the language L(R1 + R2) It is clear from figure 3. We can construct an NFA which accepts either L(R1) or L(R2) which can be represented as L(R1 + R2) as shown in figure 3. q1 which is the start state of machine M1 becomes the start state of the combined machine M and f2 which is the final state of machine M2. f2) be a machine which accepts the language L(R2) corresponding to the regular expression R2. Let M1 = (Q1.
3. 4 a 5 Step 2: The machine to accept ‘b’ is shown below. qf The regular expression corresponding to this language is ab(a+b)*.5 that the machine can either accept or any number of L(R1)s thus accepting the language L(R1)*. Here. q0 is the start state qf is the final state. 2 3 6 b 7 4 a 5 8 9 17 . 3 6 b 7 4 a 5 8 Step 4: The machine to accept (a+b)* is shown below. Step 1: The machine to accept ‘a’ is shown below.5 To accept the language L(R1)* It is clear from figure 3. Obtain an NFA which accepts strings of a’s and b’s starting with the string ab.(a) q0 q1 f1 M1 (b) Fig. 6 b 7 Step 3: The machine to accept (a + b) is shown below.
9 Generalized transition graph where r1. the regular expression can be of the form r = r1*r2 r4* (3.6 To accept the language L(ab(a+b)*) Obtain the regular expression from FA Theorem: Let M = (Q. Then there exists an equivalent regular expression R for the regular language L such that L = L(R).1) 3 Note: 1. . r3 and r4 are the regular expressions and correspond to the labels for the edges. q0. Any graph can be reduced to the graph shown in figure 3. A) be an FA recognizing the language L. Then substitute the regular expressions appropriately in the equation 3.Step 5: The machine to accept ab is shown below.2) 3. a 0 a 1 b 2 3 4 5 8 9 6 b 7 Fig. If r3 is not there in figure 3.9. 3. r2. If q0 and q1 are the final states then the regular expression can be of the form r = r1* + r1*r2 r4* (3. . Consider the generalized graph r1 q0 r 2 r q1 4 r Fig. The general procedure to obtain a regular expression from FA is shown below.9.3) 18 . The regular expression for this can take the form: r = r1*r2 (r4 + r3r1*r2)* (3. 0 a 1 b 2 Step 6: The machine to accept ab(a+b)* is shown below. 3. 2.1 and obtain the final regular expression.
it can be removed and the following FA is obtained. state q2 is the dead state. So. to reach q1 one can input any number of 0’s followed by 1 and followed by any number of 1’s and can be represented as 0*11* 19 .Obtain a regular expression for the FA shown below: 0 q0 1 0 q2 1 The figure can be reduced as shown below: 01 q0 10 It is clear from this figure that the machine accepts strings of 01’s and 10’s of any length and the regular expression can be of the form (01 + 10)* What is the language accepted by the following FA q1 1 q3 0 0. 0 q0 1 q1 1 The state q0 is the final state and at this point it can accept any number of 0’s which can be represented using notation as 0* q1 is also the final state.1 0 q0 1 q1 1 0 0. q2 1 Since.
For example. An application of regular expression in UNIX editor ed. Note that the editor ed accepts the regular expression and searches for that particular pattern in the text. As the input can vary dynamically. the final regular expression is obtained by adding 0* and 0*11*. Applications of Regular Expressions Pattern Matching refers to a set of objects with some common properties.So. In UNIX operating system. So. it is challenging to write programs for string patters of these kinds. 20 . if the command specified is /acb*c/ then the editor searches for a string which starts with ac followed by zero or more b’s and followed by the symbol c.E = 0* + 0*11* = 0* ( + 11*) = 0* ( + 1+) = 0* (1*) = 0*1* It is clear from the regular expression that language consists of any number of 0’s (possibly ) followed by any number of 1’s(possibly ). the regular expression is R. we can use the editor ed to search for a specific pattern in the text. We can match an identifier or a decimal number or we can search for a string in the text.
In other words if we have to remember n. Design of switching circuits.Jay Kant Pratap Singh Yadav Properties of regular languages • • Pumping Lemma Used to prove certain languages like L = {0n1n  n ≥ 1} are not regular. which is not possible with a finite state machine.1 0 1 4 0 1 5 0. Ex. Closure properties of regular languages Used to build recognizers for languages that are constructed from other languages by certain operations. 00*11* is not the regular expression defining L.1 State 6 is a trap state. This implies DFA has no memory to remember arbitrary ‘n’. which varies from 1 to we have to have infinite states. state 3 remembers that two 0’s have come and from there state 5 remembers that two 1’s are accepted. which has finite number of states. • Pumping Lemma for regular languages ( Explanation) Let L = {0n1n  n ≥ 1} There is no regular expression to define L. Automata for intersection of two regular languages Decision properties of regular languages – – Used to find whether two automata define the same language Used to minimize the states of DFA eg. Let L= {0212} 0 1 2 0 3 1 1 6 0. 21 .
y=an and z = . Let the start state be P1. y > 0 2. 22 .a3an and is in L.an+1 is accepted k=10 a1 . such that: 1. For all k ≥ 0. we can break w into three strings.an1 (an)k k=0 k=1 k=2 a1 . Therefore xykz = a1 .Pumping Lemma (PL) for Regular Languages Theorem: Let L be a regular language.an1 is accepted a1 . the string xykz is also in L. Then there exists a constant ‘n’ (which depends on L) such that for every string w in L such that w ≥ n. PROOF: Let L be regular defined by an FA having ‘n’ states.a2 . xy ≤ n 3. w = n ≥ n.a3 an1 . Let w = xyz where x= a1. w=xyz. Let w= a1.an is accepted a1 .a2 .an+9 is accepted and so on.
the break up of x. xy contain only a’s. To prove that L={ww is a palindrome on {a. Let n is the constant (PL Definition).. resulting string is not in L. abbbba. (Because xy ≤ n). w=xykz. Let y=l. Example 1. Since 2n+1 > n and L is regular it must satisfy PL. Then. write separate expression.1. such that w=2n. Contradiction. should belong to L. It should never be used to show that some language is regular. Let w = anban.…} Proof: Let L be regular. for all k=0. If you want to show that language is regular. where k=0. we get anl bn L. That is anl (al)k bn L.e. Hence the Language is not regular. where l > 0 (Because y > 0). Consider a word w in L. where n ≥ 1} is not regular Proof: Let L be regular.1. Let w = anbn. Consider a word w in L. 23 . General Method of proof: (i) (ii) Select w such that w n Select y such that y 1 (iii) Select x such that xy n (iv) Assign remaining string to z (v) Select k suitably to show that. Since 2n > n and L is regular it must satisfy PL.. such that w=2n+1. DFA or NFA.Uses of Pumping Lemma: . Let n is the constant (PL Definition). i. Example 2. To prove that L={ww anbn.2. y and z can be as follows from the definition of PL .2.This is to be used to show that. aba. certain languages are not regular.b}*} is not regular. L={aabaa. Put k=0.
(Because xy ≤ n). Example 3.111 . where l > 0 (Because y > 0). L={02.e.xy contain only a’s. the break up of x. for all k=0.this can not be prime if pm ≥ 2 or 1+m ≥ 2 1.1. by PL xykz L  xykz  =  xz  +  yk  = (pm) + m (pm) = (pm) (1+m) . That is anl (al)k ban L. Example 4.2..} (1+m) ≥ 2 because m ≥ 1 Limiting case p=n+2 (pm) ≥ 2 since m ≤n Let k = pm Proof: Let L be regular. 13 .1. Put k=0.. 04 . it is not a palindrome. To prove that L={ 0i2  i is integer and i >0} is not regular.} Proof: Let L be regular.17 . Let w = 0n2 where w = n2 ≥ n by PL xykz L.025 . That is.. Contradiction. y and z can be as follows from the definition of PL w=xykz.e. we get anl b an L.2. should belong to L. 2. because. where k=0. i. hence the language is not regular ..Select k = 2  xy2z  =  xyz  +  y  24 .1. i.016 . Let y=l. To prove that L={ all strings of 1’s whose length is prime} is not regular.15 . for all k = 0.09 . Let w = 1p where p is prime and  p = n +2 Let y = m. L={12.
m 0 and n<m } (ii) L={anbm  n. m 1 } (iv) L={an  n is a perfect square } (v) L={an  n is a perfect cube } b) Apply pumping lemma to following languages and understand why we cannot complete proof (i) L={anaba  n 0 } (ii) L={anbm  n. m 0 } 25 . m 0 and n>m } (iii)L={anbmcmdn  n. Exercises for students: a) Show that following languages are not regular (i) L={anbm  n.= n2 + Min 1 and Max n Therefore n2 <  xy2z  ≤ n2 + n n2 <  xy2z  < n2 + n + 1+n n2 <  xy2z  < (n + 1)2 adding 1 + n ( Note that less than or equal to is replaced by less than sign) Say n = 5 this implies that string can have length > 25 and < 36 which is not of the form 0i2.
a4. Ex1. 2.a5b5. a5b5} RE=ab(ab)* Closure Under Complementation Theorem : If L is a regular language over alphabet S. The closure (star) of a regular language is regular. Ex1.a6. 7. The reversal of a regular language is regular.L is also a regular language.a4. then so is L M. 9.a4. A homomorphism (substitution of strings for symbols) of a regular language is regular.} L2={ab. a3b3.} L2={a2.Closure Properties of Regular Languages 1. 4. L1={ab.} * L1={e. L1={a. The inverse homomorphism of a regular language is regular Closure under Union Theorem: If L and M are regular languages. The complement of a regular language is regular.a5.a3. then L = * .a3. 5. a4b4.a5.} RE=(aa)* 26 . a4b4. L1={a.a6.a2. a3b3. The intersection of two regular languages is regular.} L1L2 = {ab.a2. 3. The concatenation of regular languages is regular.a2b2. 6. 8.} RE=a(a)* Ex2.} L1L2 = {a.a3 b3.a3. The difference of two regular languages is regular. The union of two regular languages is regular. a2 b2.
The complement of L(A) is therefore all string of 0’s and 1’s that do not end in 01 27 . A that accepts all and only the strings of 0’s and 1’s that end in 01.Ex2. That is L(A) = (0+1)*01. Consider a DFA.
} L1L2 = {a2.a3.If L is a regular language over alphabet . B is exactly like A. . q0. That is.a5b5. but the accepting states of A have become nonaccepting states of B. then so is L M.a6. L = * .a7b7} L2={a2 b2. F).Theorem: . then w is in L(B) if and only if ^ ( q0. 28 . q0.a4.a3b3.a4. L1={a. . where B is the DFA (Q.a6. then. A=(Q. . w) is in QF. .} L2={a2.a5.} RE=aa(aa)* Ex2 L1={ab. a6b6. Consider a DFA that accepts all those strings that have a 1.a4. a4b4.Let L =L(A) for some DFA. Consider a DFA that accepts all those strings that have a 0. Ex1. QF). Then L = L(B).L is also a regular language Proof: . Closure Under Intersection Theorem : If L and M are regular languages.a6.} L1L2 = RE= Ex3. The product of above two automata is given below. and vice versa.a2. which occurs if and only if w is not in L(A).
DFA for L1 L2 = (as no string has reached to final state (2. Then pr represents only the initial condition. Ex 4 (on intersection) Write a DFA to accept the intersection of L1=(a+b)*a and L2=(a+b)*b that is for L1 L2.4)) Ex5 (on intersection) Find the DFA to accept the intersection of L1=(a+b)*ab (a+b)* and L2=(a+b)*ba (a+b)* that is for L1 L2 DFA for L1 L2 29 . The accepting state qs represents the condition where we have seen both 0’s and 1’s. in which we have seen neither 0 nor 1. while state ps represents the condition that we have seen only 1’s. Then state qr means that we have seen only once 0’s.This automaton accepts the intersection of the first two languages: Those languages that have both a 0 and a 1.
a6. M = (Q. If L is regular we can show that LR is also regular. having only one final state. Proof. 10.111}. L={001. . Ex.01} LR={100. As L is regular it can be defined by an FA.a3. then so is L – M.a7} RE=a(a)* Reversal Theorem : If L is a regular language.a5.} L2={a2.} L1L2 = {a. . If there are more than one final states. That is LR = {100.Closure Under Difference Theorem : If L and M are regular languages.111. Let L = {001.a5. q0.10. L1={a. 30 . be a language over ={0. 111}.111.1}.a7. so is LR Ex.a3.01.10} To prove that regular languages are closed under reversal. we can use .01. F).transitions from the final states going to a common final state. LR is a language consisting of the reversals of the strings of L.a4.
a) > p Since MR is derivable from M. R . That is L = {ab.aaab.a)> q. 31 . MR = (QR. Where QR = Q. }. R.FR=q0.FR) defines the language LR. Swap initial and final states. aab.q0R. 2. The proof implies the following method 1. q0R=F. LR is also regular. R = . iff (q. The FA is as given below The FA for LR can be derived from FA for L by swapping initial and final states and changing the direction of each edge. Create a new start state p0 with transition on to all the accepting states of original DFA Example Let r=(a+b)* ab define a language L. Reverse all the transitions. 3. bab. and R (p.Let FA. It is shown in the following figure.
Ex.Jay Kant Pratap Singh Yadav Homomorphism A string homomorphism is a function on strings that works by substituting a particular string for each symbol. Theorem : If L is a regular language over alphabet . and L is a regular language over T. h applied to the string 00110 is ababccab L1= (a+b)* a (a+b)* h : {a. then h (L) is also regular. and h is a homomorphism on . 32 . b} {0. The function h defined by h(0)=ab h(1)=c is a homomorphism. then h1 (L) is also a regular language. 1}* Resulting : h1(L) = (01 + 11)* 01 (01 + 11)* h2(L) = (101 + 010)* 101 (101 + 010)* h3(L) = (01 + 101)* 01 (01 + 101)* Inverse Homomorphism Theorem : If h is a homomorphism from alphabet S to alphabet T.
Testing Emptiness of Regular Languages Suppose R is regular expression. Therefore the bound on the running time is O(n3s) where s is the number of states the DFA actually has. Then L(R) is empty if and only if either L(R1) or L(R2) is empty. Conversion from NFA to NFA takes O(n3) time. Is a particular string w in the described language? 3. Testing Emptiness of Regular Languages Suppose R is regular expression. then 1. Then h1(L) is the language of regular expression (ba)*. Let h be the homomorphism defined by h(a)=01 and h(b)=10. R=(R1) Then L(R) is empty if and only if L(R1) is empty since they are the same language. R= R1R2. Computing Closure of n states takes O(n3) time.Let L be the language of regular expression (00+1)*. Do two descriptions of a language actually describe the same language? This question is often called “equivalence” of languages. Is the language described empty? 2. The running time of NFA to DFA conversion including transition is O(n3 2n). 3. It always includes at least 4. then 1. Then L(R) is empty if and only if both L(R1) and L(R2) are empty. Converting Among Representations Converting NFA’s to DFA’s Time taken for either an NFA or NFA to DFA can be exponential in the number of states of the NFA. Automaton to Regular Expression Conversion For DFA where n is the number of states. Computation of DFA takes O(n3) time where number of states of DFA can be 2n. R = R1 + R2. conversion takes O(n34n) by substitution method and by state elimination method conversion takes O(n3) time. R=R1* Then L(R) is not empty. Decision Properties of Regular Languages 1. 2. 33 3 . Then L(R) is empty if and only if both L(R1) and L(R2) are empty.Ex. DFA to NFA Conversion Conversion takes O(n) time for an n state DFA. R = R1 + R2. If we convert an NFA to DFA and then convert the DFA to a regular expression it takes the time O(n34n 2n) Regular Expression to Automaton Conversion Regular expression to NFA takes linear time – O(n) on a regular expression of length n.
If the representation of L is a Regular Expression of size s. Testing Membership in a Regular Language Given a string w and a Regular Language L. Two states p and q are said to be distinguishable. NFA has s states. running time of this algorithm is O(ns2) If the representation is . If L is represented by a DFA. has 2 stages. Examples: 1. w. Simulation of the above takes O(ns2) time on an input w of length n Minimization of Automata ( Method 1) Let p and q are two states in DFA. R= R1R2.NFA. After a pass in which no new pairs are marked.closure has to be computed. If DFA ends in accepting state the answer is ‘Yes’ . if there is at least one string.q) are distinguishable. such that one of ^ (p. mark each pair of which exactly one element is in F. a . a6. a4.b}. .a) = p. then processing of each input symbol . (s. …. in O(s) time. 3. and (p. Fig 2. if w is of length n.w) is accepting and the other is not accepting. mark any pair (r.2. a2. It always includes at least 4.q) is already marked. we can convert to an NFA with almost 2s states.q) for which p ≠ q. beginning in start state. The FA is shown in Fig 1. Make a sequence of passes through these pairs. each of which requires O(s2) time. is w in L. 34 . Then L(R) is empty if and only if either L(R1) or L(R2) is empty. On each subsequent pass. else it is ‘no’. R=(R1) Then L(R) is empty if and only if L(R1) is empty since they are the same language.a) = q.w) and ^ (q. Let L = {. Our goal is to understand when p and q (p ≠ q) can be replaced by a single state. simulate the DFA processing the string of input symbol w.} be a regular language over = {a.s) if there is an a for which (r. gives the list of all unordered pairs of states (p. On first pass. R=(R1)* Then L(R) is not empty. The marked pair (p. This test takes O(n) time If the representation is NFA. stop. Algorithm 1: List all unordered pair of states (p.q) with p ≠ q.
).2) and (2. In pass 2 no boxes are marked because.a) and (3. (1.q) with p ≠ q 35 . (1.110. This implies state 1 and 3 are equivalent and can not be divided further. (Method1): Let r= (0+1)*10.3) (.3) (2. (1. }. The resultant FA is shown is Fig 3. where is a nonfinal state.). That is (1.010.3) are equivalent and can replaced by a single state A.3) are marked in the first pass according to the algorithm 1. This gives us two states 2. Minimal Automata corresponding to FA in Fig 1 Minimization of Automata (Method 2) Consider set {1.3) (.A. (. then L(r) = {10.2) and (1.2). This implies that (1.a) 2. Fig 3.The boxes (1. The FA is given below Following fig shows all unordered pairs (p.b) and (3. That is (1.3) where and 3 are non final states.b) .3}. Example 2.00010.
5. and the pair (6. 6 The transitions of fig 4 are mapped to fig 6 as shown below Example 2.3) (4.7) and (2. since (5.4).2) (6. and 7 can be replaced by the single state 357. similarly it can be easily shown for the pairs (4. The resultant minimal FA is shown in Fig.5) (1. 2. (Method1): (2.4) was marked on pass 1. and 4 can be replaced by a single state 124 and states 3.2) is one of these. For example (5. The pairs marked 2 are those marked on the second pass.The pairs marked 1 are those of which exactly one element is in F. They are marked on pass 1.6) this implies that 2 and 3 belongs to different group hence they are split in level 2.5) and so on. From this we can make out that 1. 36 .
P) Where A represents the set of five productions 1. P. G is represented by four components that is G= (V. T. 3. {0. where V is the set of variables. 2. 1}. T the terminals. starting with the string E. A. 4.Jay Kant Pratap Singh Yadav Lecture Notes for Class 4 (21/4/2005) Context Free Grammar Context Free grammar or CGF. P the set of productions and S the start symbol. S). 5. Example: The grammar Gpal for palindromes is represented by Gpal = ({P}. Derivation using Grammar P P0 P1 P0P0 P1P1 Example 1: Leftmost Derivation The inference that a * (a+b00) is in the language of variable E can be reflected in a derivation of that string. Here is one such derivation: E E * E I * E a * E a * (E) a * (E + E) a * (I + E) a * (a + E) a * (a + I) a * (a + I0) a * (a + I00) a * (a + b00) 37 .
Tree E* There is a rightmost derivation that uses the same replacements for each variable. we can describe the same derivation by: E E * E E *(E) E * (E + E) E * (E + I) E * (E +I0) E * (E + I00) E * (E + b00) E * (I + b00) E * (a +b00) I * (a + b00) a * (a + b00) We can also summarize the leftmost derivation by saying E a * (a + b00).Leftmost Derivation . Thus. although it makes the replacements in different order.Tree Example 2: Rightmost Derivations The derivation of Example 1 was actually a leftmost derivation. or express several steps of the derivation by expressions such as E a * (E). This rightmost derivation is: 38 . Rightmost Derivation .
E * (I + E) is a sentential form.T. denoted by L(G). is the set of terminal strings that have derivations from the start symbol. P. L(G) = {w in T  S w} Sentential Forms Derivations from the start symbol produce strings that have a special role called “sentential forms”.S)(a. the language of G. then is a left – sentential form.S) is a CFG.S)(a.a)(S. T. since there is a derivation E E * E E * (E) E * (E + E) E * (I + E) 39 . and if S . Note that the language L(G) is those sentential forms that are in T*.a) S>(L)a L>L. then is a right – sentential form.a)(a.SS Leftmost derivation S(L)(L.a) Rightmost Derivation S(L)(L.E E * E E * (E) E * (E + E) E * (E + I) E * (E + I0) E * (E + I00) E * (E + b00) E * (I + b00) E * (a + b00) I * (a + b00) a * (a + b00) This derivation allows us to conclude E a * (a + b00) Consider the Grammar for string(a+b)*c EE + T  T T T * F  F F ( E )  a  b  c Leftmost Derivation ETT*FF*F(E)*F(E+T)*F(T+T)*F(F+T)*F (a+T)*F (a+F)*F (a+b)*F(a+b)*c Rightmost derivation ETT*FT*cF*c(E)*c(E+T)*c(E+F)*c (E+b)*c(T+b)*c(F+b)*c(a+b)*c Example 2: Consider the Grammar for string (a. then any string in (V T)* such that S is a sentential form. That is if G = (V.S)(L. S) is a CFG. that is they consist solely of terminals. If S .S)(S.a) The Language of a Grammar If G(V. For example.P.
ambiguity implies the existence of two or more left most or rightmost derivations.E.However this derivation is neither leftmost nor rightmost.I}. EE*EI*Ea*E Additionally. As an example of a left – sentential form. and productions. 40 .*. T={a. E(E).c. EE+E. Ex:Consider the grammar G=(V. Alternatively.b. Iabc Consider two derivation trees for a + b * c.+. since at the last step.T. Ambiguity A context – free grammar G is said to be ambiguous if there exists some w L(G) which has at least two distinct derivation trees. EE*E. with the leftmost derivation.)}. the derivation E E * E E * (E) E * (E + E) Shows that E * (E + E) is a right – sentential form.P) with V={E. consider a * E. the middle E is replaced.(. EI.
Now unambiguous grammar for the above Example: ET. TF. FI. TT*F. Iabc Inherent Ambiguity A CFL L is said to be inherently ambiguous if all its grammars are ambiguous Example: Condider the Grammar for string aabbccdd SAB  C A aAb  ab BcBd  cd C aCd  aDd D>bDc  bc Parse tree for string aabbccdd 41 . EE+T. F(E).
Id : ‘a’ {…} ’b’ {…}  Id ‘a’ {…} Id ‘b’ {…} Id ‘0’ {…} Id ‘1’ {…} . 42 {…} {…} {…} {…} .Applications of Context – Free Grammars • • • • Parsers The YACC Parser Generator Markup Languages XML and Document typr definitions The YACC Parser Generator EI  E+E  E*E  (E) Ia  b  Ia  Ib  I0  I1 Exp : Id  Exp ‘+’ Exp  Exp ‘*’ Exp  ‘(‘ Exp ‘)’ .
XML and Document type definitions. 1. AE1,E2. ABC BE1 CE2 2. AE1  E2. AE1 AE2 3. A(E1)* ABA A BE1 4. A(E1)+ ABA AB BE1 5. A(E1)? A AE1 EXERCISE QUESTIONS 1) Design contextfree grammar for the following cases a) L={ 0n1n  n≥l } b) L={aibjck i≠j or j≠k} 2) The following grammar generates the language of RE 0*1(0+1)* S AB A 0A B 0B1B Give leftmost and rightmost derivations of the following strings a) 00101 b) 1001 c) 00011 43
3) Consider the grammar S aSaSbS Show that deviation for the string aab is ambiguous 4) Suppose h is the homomorphism from the alphabet {0,1,2} to the alphabet { a,b} defined by h(0) = a; h(1) = ab & h(2) = ba a) What is h(0120) ? b) What is h(21120) ? c) If L is the language L(01*2), what is h(L) ? d) If L is the language L(0+12), what is h(L) ? e) If L is the language L(a(ba)*) , what is h1(L) ?
44
%{ #include <stdio.h> %} %token ID id %% Exp : id { $$ = $1 ; printf ("result is %d\n", $1);}  Exp ‘+’ Exp {$$ = $1 + $3;}  Exp ‘*’ Exp {$$ = $1 * $3; }  ‘(‘ Exp ‘)’ {$$ = $2; } ; %% int main (void) { return yyparse ( ); } void yyerror (char *s) { fprintf (stderr, "%s\n", s); } %{ #include "y.tab.h" %} %% [09]+ {yylval.ID = atoi(yytext); return id;} [ \t \n] ; [+ * ( )] {return yytext[0];} . {ECHO; yyerror ("unexpected character");} %%
Example 2:
%{ #include <stdio.h> %} %start line %token <a_number> number %type <a_number> exp term factor %% line : exp ';' {printf ("result is %d\n", $1);} ; exp : term {$$ = $1;}  exp '+' term {$$ = $1 + $3;}  exp '' term {$$ = $1  $3;} term : factor {$$ = $1;}  term '*' factor {$$ = $1 * $3;}  term '/' factor {$$ = $1 / $3;} ; 45
return number.factor : number {$$ = $1.} . } void yyerror (char *s) { fprintf (stderr.}  '(' exp ')' {$$ = $2.} .] {return yytext[0].} %% Markup Languages Functions •Creating links between documents •Describing the format of the document Example The Things I hate 1. People who drive too slow In the fast lane HTML Source <P> The things I <EM>hate</EM>: <OL> <LI> Moldy bread <LI>People who drive too slow In the fast lane </OL> 46 . Moldy bread 2.tab. [+*/().a_number = atoi(yytext). {ECHO. "%s\n". yyerror ("unexpected character"). s). %% int main (void) { return yyparse ( ).} [ \t\n] .h" %} %% [09]+ {yylval. } %{ #include "y.
AE1.E2. A(E1)* ABA A BE1 4. A(E1)+ ABA AB BE1 5.HTML Grammar •Char •Text •Doc •Element aA… e  Char Text e  Element Doc Text  <EM> Doc </EM> <p> Doc  <OL> List </OL> … ListItem <LI> Doc List e  ListItem List Start symbol 5. ABC BE1 CE2 3. 6. AE1 AE2 4. XML and Document type definitions. A(E1)? A AE1 EXERCISE QUESTIONS 1) Design contextfree grammar for the following cases a) L={ 0n1n  n≥l } b) L={aibjck i≠j or j≠k} 3) The following grammar generates the language of RE 0*1(0+1)* S AB A 0A B 0B1B Give leftmost and rightmost derivations of the following strings 47 . 2. AE1  E2.
h(1) = ab & h(2) = ba a) What is h(0120) ? b) What is h(21120) ? c) If L is the language L(01*2). what is h(L) ? d) If L is the language L(0+12).a) 00101 b) 1001 c) 00011 3) Consider the grammar S aSaSbS Show that deviation for the string aab is ambiguous 4) Suppose h is the homomorphism from the alphabet {0.1. what is h1(L) ? 48 .2} to the alphabet { a.b} defined by h(0) = a. what is h(L) ? e) If L is the language L(a(ba)*) .
one symbol at a time. FSC does some state transitions and does some operations to the stack content. So. Similarly for Contextfree Languages. F).Session by Jay K. for equating with number of b’s found. or some thing could be pushed into the stack and could be popped out of the stack. q0. . has the form : Q X ( {}) X finite subsets of Q X * q0 Q is the initial state. PDA is more powerful than FA. and F Q is a set of final states. is a transition function. 49 . Stack could be kept unchanged. Figure 1 shows a diagrammatic representation of PDA. Based on the input symbol. z. is a the input alphabet. the contextfree languages (CFL) have corresponding machines called pushdown automata (PDA). where Q is a finite set of states. Stack functions as the required memory. An FA cannot recognize the language anbn. current state and the top symbol on the stack. . The Finite State Control (FSC) reads inputs. Regular expressions are generators for regular languages and Finite Automata’s are recognizers for them. Pratap Singh Yadav: Introduction to Push Down Automata Just as finitestate automata correspond to regular languages. a PDA is an NFA with a stack. PDA is NFA with an added memory. is the stack alphabet. Context Free Grammars (CFG) are generators and Pushdown Automata (PDA) are recognizers. because FA does not have any memory to remember the number of a’s it has already seen. n 0. Input Tape Finite State Control stack Figure 1: PushDown Automaton Formal Definition: A nondeterministic pushdown automaton or npda is a 7tuple M = (Q. . z is the stack start symbol.
2. and stack top Z0. If current state is q0. The following transitions are possible: 1. a) = {(q0. and replace the symbol ‘t’ on top of the stack with the symbol ‘u’ ”. input symbol and stack symbol. . move to q2 the final state. move from state P to state Q.Transition function: for any given state. stack top is a. 5. Z0) = {(q2. u) Basically. a. then stack is unchanged. b. q2}. a) = {(q1. then stack is popped If u= wx. and symbol on input tape is at . Z0. if. b}. it has the form: (P. b. If current state is q0. stack top is a. If current state is q1. t and P and Q are states. then stay in q0 and push ‘a’ to the stack. a. Means “read the symbol ‘a’ from the input. q0. b. move to state q1 and pop the top symbol of the stack. Example 1: Construct PDA for the language L= {anbn  a. If current state is q0. aZ0)} δ(q0. a . 3. input tape symbol is ‘a’. and input tape symbol is a.Y) = for all other possibilities 50 . If current state is q1. {a. then move to q2 the final state. a. then t is replaced with x and w is pushed into the stack. and stack top is Z0.e. t) (Q. )} δ(q0. i. input tape symbol is b. So we can define PDA as M = ({q0. input tape symbol is and stack top is Z0. )} δ(q1. If current state is q0. stay in q1 and pop the top symbol of the stack 6.x. )} δ(q1. If u = . δ. where δ is defined by following rules: δ(q0. gives a new state and stack symbol. a) = {(q1. input tape symbol is b. aa)} δ(q0. and stack top is a. {q2}). stay in q0 and push ‘a’ to the stack. 4. Z0) = {(q2. Z0}. If u = t. q1. b Σ n 0}. . )} δ(q. {a. Start at state q0 and keep Z0 in the stack. Z0) = {(q0.
where 1. Transition diagram of PDA is generalization of transition diagram of FA. 1. u is the stack contents. Doubly circled states are final states 4.a/ a. Arc corresponds to transitions of PDA. So u Γ* Moves of A PDA: Let the symbol "" indicates a move of the nPDA. There are two types of moves possible for a PDA. Z0 / Figure 2: Transition diagram Instantaneous Description: Instantaneous Description or configuration of a PDA describes its execution status at any time. with the leftmost symbol at the top of the stack. Arrow labeled Start indicates start state 3. )…} is an arc labeled (a. written as a string.a/ Start q0 q1 . Z0/aZ0 a. 2.Graphical Notation of PDA: To understand the behavior or PDA clearer.a/aa b. a. b. Node corresponds to states of PDA 2. Z0 / q2 . w is the unread part of the input string. w Σ* 3. X) = {(p. If δ(q. q is the current state of the automaton. 1. moves automaton to state q and replaces the stack top with . The transition diagram for the above example PDA is given in Figure 2. w. u). X/) from state q to state p means that an input tape head positioned at symbol a and stack top with X. the transition diagram of PDA can be used. Move by consuming input symbol 51 . Instantaneous Description is a represented by a triplet (q.
}. So the string is accepted. b. a) = {(q0. )} . a) = {(q0. )} PDA reached a configuration of (q2. .(q0. aZ0) as per transition rule δ(q0. bb. a) = {(q1. Z0) = {(q2. . aaZ0) as per transition rule δ(q0.(q1. abb. aa)} . an input symbol ‘a’ is consumed from the input string aW.Suppose that (q1. Z0) as per transition rule δ(q0. Z0) = {(q0. aabb. .(q0. aaZ0) as per transition rule δ(q0.There is no defined move. a. Z0) . a. Z0) . aZ0) as per transition rule δ(q0. Z0) . leaving yZ on the stack. a. b) Moves for the input string aaabb: (q0.(q1. a. aaabb.(q0.(q1. a) = {(q1. The input tape is empty. . aabb. aabbb. Then the following move by consuming an input symbol is possible: (q1..(q1. )} . abb. a. a) = {(q1. aZ0)} . b. abbb. a. aZ0) as per transition rule δ(q0. aaZ0) as per transition rule δ(q0. a. b. Z0) = {(q0. y).(q0. Z0) = {(q0. aZ0)} . W.(q1. This notation says that in moving from state q1 to state q2. . aaaZ0) as per transition rule δ(q0.. bb. aW. The above example PDA with a few example input strings. . a) = {(q1. )} .(q0. and the symbol ‘x’ at the top (left) of the stack xZ is replaced with symbol ‘y’. bb. a) = {(q0. bbb. a. a) = {(q1. the moves are given below: a) Moves for the input string aabb: (q0. So the automaton stops and the string is not accepted. aZ0)} .(q1.(q0. a) = {(q0. aZ0) as per transition rule δ(q0. b. aZ0) as per transition rule δ(q0. a) = {(q1. xZ) . aZ0) as per transition rule δ(q0. b.). aaZ0) as per transition rule δ(q0. x) = {(q2. where W indicates the rest of the input string following the a. )} . )} 52 . aa)} . b.(q2. Z0) as per transition rule δ(q0. )} . and Z indicates the rest of the stack contents underneath the x. b. c) Moves for the input string aabbb: (q0.(q0. aa)} . aa)} . yZ). b.) as per transition rule δ(q1. stack is empty and PDA has reached a final state. b.(q2.
0) ={(c. 2.. yZ). Z) ={(d. )} (b. 0) ={(b. {0. This move is usually used to represent nondeterminism.There is no defined move.used to represent zero or more moves of PDA. The relation * is the reflexivetransitive closure of . {0. the tape head position will not move forward. b. and the symbol ‘x’ at the top (left) of the stack xZ is replaced with symbol ‘y’. So the automaton stops and the string is not accepted.1}. .. For the above example. Solution: The transition diagram could be given as figure 3. . In this move. x) = {(q2. 0) ={(c. 0Z)} (b. 0. where is given by: (a. xZ) .). Example 2: Design a PDA to accept the set of all strings of 0’s and 1’s such that no prefix has more 1’s than 0’s. 0)} (b. 0) ={(d. . leaving yZ on the stack. Figure 3: Transition diagram M = ({a. aW. Z) ={(b. 0. . an input symbol ‘a’ is not consumed from the input string aW. Z) ={(b. 1. Then the following move without consuming an input symbol is possible: (q1.(q2. 0Z)} 53 .1.}. )} (c. Z)} (c. 00)} (b. a. Z. . (q0.Z}. y). 0) ={(b. . c. aW. 00)} (c. 1. This notation says that in moving from state q1 to state q2. 0. Z0) * (q2. d}. ..move Suppose that (q1. {d}). aabb. 0.
So the input string is accepted by the PDA.Z) no move.(b.011. 0Z) . .10110. . Moves for the input string 0010110 is given by: (a. Z)} For all other moves. 010110. 10. 00Z) (c. 0Z).11. Z) .(b. Show the moves for abbaba 3. b Σ. 0Z) . 00Z) . 0010110. . 0Z) .0Z) . For the language L = {anb2n  a. Moves for 011 (a.(d.1. 0) ={(d. PDA stops. b}*.(b. (c.(c.Z) . . so PDA stops Exercises: Construct PDA: 1. 110. 0)} (c.(b. 0110. (consider any type of parentheses) 54 . 0. Z) .(c. n 0} 4.(c. Accepting the set of all strings over {a.(b. 0Z) . Z) ={(d. Accepting the language of balanced parentheses. b} with equal number of a’s and b’s. For the language L = {wcwR  w {a. c Σ } 2.
0) = {(q0. 1. 0110. PDA enters a final state. )}. 1. 00)} δ(q0. 11)} δ(q0. In the string 0110. the appropriate PDA could be by final state. Acceptable input strings are like 00. The PDA could be constructed as below. Usually the languages that a PDA accept by final state and PDA by empty stack are different. Z0) * (q. {0. 0) = {(q0. Languages of PDA There are 2 ways of accepting an input string PDA a. It is possible to covert a PDA accept by final state to another PDA accept by empty stack and also the vice versa. F) be a PDA. . stack of the PDA will be empty. {0. 0Z0)} δ(q0. Σ. q2}. Accept by empty stack After consuming the input. 0. 101101. and 110011. q2). 0. Both methods are equivalent. The current state could be final or nonfinal state. q0. 1Z0)} δ(q0. Accept by Final state: Let P = (Q. q1. where δ is defined by: δ(q0. 1}. the difficulty is how to decide the middle of the input string? The 3rd 1 can be part of w or can be part of wR. 1) = {(q0. Z0. Z0) = {(q0. 1) = {(q0. the language for even length palindrome. For example the language {L = anbm  n m}. 1. M = ({q0. 1. Γ. 1111. the stack may not be empty.Session by Jay K. Accept by Final state After consuming the input. δ. Pratap Singh: Finite Automata & Formal Languages Push Down Automata Languages of PDA 1. 10)} δ(q0. Z0. Z0)} 55 . where q F and Γ* Example: L = {wwR  w is in (0 + 1)*}. w. the language accepted by P by the final state is {w  (q0. 01)} δ(q0. q0. Z0) = {(q1. . Then L(P). b. After consuming the input. Z0) = {(q0. δ. The content of the stack is irrelevant. 0.Z0}.
. 1)} δ(q1.0/ 1. Z0) = {(q2. wwR. 0) = {(q1. )} δ(q1.0/0 .δ(q0. . 1. 1) = {(q1. Z0) . 0.1/1 . 0) = {(q1. . 1) = {(q1. Z0) The transition diagram for the PDA is given in figure 1. Z0) * (q0.0/00 1. Z0 / Z0 q2 Figure 1: Transition Diagram for L = {wwR} The moves of the PDA for the input string 101101 are given figure 2. Z0/0Z0 1.(q2.1/11 0.0/ 10 0. 0. wR. Z0)} (q0. Z0/1Z0 0. )} δ(q1. wR. 0)} δ(q0. .1/ 01 1.1/ . Z0/ Z0 q0 q1 .(q1. wRZ0) . wRZ0) * (q1. Figure 2: moves of PDA for string 101101 56 . .
Z0) = {(q2. Z0/0Z0 1. )} to get accept by empty stack. Z0/Z0 q . 0/00 0. 0. Σ. Z0/Z0 s p . Z0/Z0 1. then there is a PDA PF such that L = L(PF) Proof: p0 .Z0/ r j Figure 3: transition diagram of 0i1  0 i j 2. Conversion between the two forms: a. Z0) * (q. From Empty Stack to Final State: Theorem: If L = N(PN) for some PDA PN= (Q. The set of accepting states are irrelevant. This example also shows that L(P) = N(P) Example: j Construct PDA to accept by empty stack for the language L={0i1  0 i j} The transition diagram for the PDA is given in Figure 3. 57 . w. X0/Z0X0 PN q0 .Accept by empty stack: Let PDA P = (Q. . Γ. X0/ (add this transition from all states of PN to new state Pf) Pf Figure 4: PF simulates PN The method of conversion is given in figure 4. Z0)} give δ(q1. . . Γ . 0/ 1. Z0). where q Q Example: Construct PDA to accept by empty stack for the language L = {wwR  w is in (0 + 1)*} Instead of the transition δ(q1. We define the language accepted by empty stack by N(P) = {w  (q0. 0/ 1. q0. Z0/Z0 . δ.Z0/Z0 1. q0. Z0). Z0) = {(q2. )}. δN. Σ.
We use a new symbol X0. Γ{X0}. a.. a Σ or a = and y Γ. Also add a new start state p0 and final state p for PN. F). Also add a new start state p0 and final state pf for PF. δN. X0) PF (p0. same for both PN and PF. The moves of PF to accept a string w can be written like: (p0. initially change the stack start content from Z0 to Z0X0. X0) = {(q0. ) b. pf}. . same for both δN(q . Z0X0) *PN (q. which must be not symbol of Γ to denote the stack start symbol for PF. . w. where δF is defined by δF(p0. ) 58 . . X0/Z0X0 PF q0 . Γ{X0}. . / Figure 5: PN simulates PF The method of conversion is given in figure 5. Σ. Let PF = (Q{p0. To avoid PF accidentally empting its stack. p0. w. X0). The moves of PN to accept a string w can be written like: (p0. δF(q. q0. same for both δN(p . / / p . X0) PN (q0. {Pf}). X0) = {(Pf. a. to pop the remaining stack contents. p}. a. y) = {(p . y Γ or y = X0. y). a. y Γ or y = X0 . . Γ. y) a Σ or a = and y Γ. . Z0 X0)} to push X0 to the bottom of the stack δF(q. y) = δN(q. Z0X0) *PF (q.(Pf . p0. y) = {(p. X0) = {(q0.(p. )}. X0) . w. Z0. From Final State to Empty Stack: Theorem: If L = L(PF) for some PDA PF= (Q. δF. )}. . Let PN = (Q{p0. X0. . w. . q F. then there is a PDA PN such that L = N(PN) Proof: p0 . y) = δF(q . X0) . where δN is defined by: δN(p0. )} to accept the string by moving to final state. . Σ. Σ. δF. Z0X0)} to change the stack content initially δN(q .
{0.0) = {(q1. Z}. )} Exercises: Design nPDA to accept the language: 1. j 0} 3. 0) = {(q0. δ’. j. )} δ(q2. {0. Solution: PDA by final state is given byM = ({q0. q0. Z) = {(q0.0) = {(q0. 1. 1. {aibjck  i. PDA by empty stack is given by M = ({q0. . q0. Z)} δ(q0. j 1} 59 . 1}. Also convert the PDA to accept by empty stack. 0. q1. .0) = {(q2.Example: Construct PDA to accept by final state the language of all strings of 0’s and 1’s such that number of 1’s is less than number of 0’s. )} δ(q2. . {aibi+jcj  i 0. Z) = {(q0. 1. Z). {q1}). Z. . )} δ(q0.0) = {(q2. 0.Z) = {(q1. 0Z)} δ(q0.Z) = {(q2. 11)} δ(q0. {0. 1) = {(q0. 1. PDA stops. . where δ is given by: δ(q0. )} δ(q1. 0)} For all other moves.Z) = {(q2.Z}. where δ’ is the union of δ and the transitions given below: δ(q1. 1Z)} δ(q0. {aibjci+j  i. 1}. δ. 00)} δ(q0.1) = {(q0. . 1. q2}. 0. k 0 and i = j or i = k} 2. )} δ(q0. q1}. {0.
At the end (case u = w) the stack will be empty. a. δ. Language accepted by PDA by final state 3. Most transitions are on . New transitions are added. q. b)  A b is a production of Q} 2. δ(q. Let G = (V. T. The PDA which accepts L(G) by empty stack is given by: P = ({q}. For every intermediate sentential form uA in the leftmost derivation of w (initially w = uv for some v). S) be a CFG. The representation is shown in figure 1. There is only one state in the new PDA. M will have A on its stack after reading u. all the rest of the information is encoded in the stack. each one corresponding to terminals of G. For each terminal a. T. V T. . include a transition δ(q. Q. S) where δ is defined by: 1. Context Free Language defined by CFG 2. Equivalence of PDA and CFG The aim is to prove that the following three classes of languages are same: 1. a) = {(q. For each variable A include a transition.Finite Automata & Formal Languages Push Down Automata Equivalence of PDA and CFG I. Language accepted by PDA by empty stack It is possible to convert between any 3 classes. CFG PDA by empty stack PDA by Final state Figure 1: Equivalence of PDA and CFG From CFG to PDA: Given a CFG G. )} 60 . The stack symbols of the new PDA contain all the terminal and nonterminals of the CFG. A) = {(q. one for each production. we construct a PDA P that simulates the leftmost derivations of G.
1}. Z). q. {0.CFG to PDA conversion is another way of constructing PDA. S). X) = {(q. (q. A)} δ(q. 0) = {(q. where V = {S. (q. Γ. For all states p. S). G has productions S [q0Z0 p] 2. q0. Z) = {(q. 1. e}. q. A. 0. Let the PDA P = ({q}. 1A0). e.Z}. δ. where δ is given by: δ(q. R. 0S1). k can be 0 or any number and r1r2 …rk are list of states. where p. A) = {(q. δ(q. i.a. )} Solution: Equivalent productions are: S [qZq] 61 . XZ)}. . Example: Convert the grammar with following production to PDA accepted by empty stack: S 0S1  A A 1A0  S  Solution: P = ({q}. δ. Σ. S) = {(q. An equivalent CFG is G = (V. q Q and X Γ. {i. . Y1Y2…Yk)} where a Σ or a = . productions of R consists of 1. Z) = {(q. G has productions [qXrk] a[rY1r1] [r1Y2r2] … [rk1Ykrk] If k = 0 then [qXr] a Example: Construct PDA to accept ifelse of a C program and convert it to CFG.X) = {(r. First construct CFG. where δ is given by: δ(q. δ. S}. )} δ(q. Σ. )} From PDA to CFG: Let P = (Q. and then convert CFG to PDA. . )} and δ(q. {0. (This does not accept if –if –elseelse statements). Z0) be a PDA. )} δ(q. [pXq]}. S). 1. (q. Let δ(q. {X. 1) = {(q.
AiBA  . X) = {(p. 0. B e}. XZ)} [qZq] 1[qXq][qZq] [qZq] 1[qXp][pZq] [qZp] 1[qXq][qZp] [qZp] 1[qXp][pZp] For δ(q. )} [qXq] For δ(q. )} δ(p. Z) = {(q. 0. . S) Example: Convert PDA to CFG. 1.[qZq] i[qXq][qZq] [qXq] e [qZq] If [qZq] is renamed to A and [qXq] is renamed to B. {X. XZ)} δ(q. Z)} Solution: Add productions for start variable S [qZq]  [qZp] For δ(q. X)} δ(p. q. A. X) = {(q. {SA. PDA is given by P = ({p. Z)= {(q.1}. 1.q}. X) = {(p. X) = {(q. then the CFG can be defined by: G = ({S. e}. XX)} δ(q. X) = {(p. 1. {0. 0. X)} [qXq] 0[pXq] [qXp] 0[pXp] 62 . )} δ(q. Z) = {(q. 1. X)= {(q. {i. 1. Z)).Z}. . X) = {(q. B}. δ. Transition function δ is defined by: δ(q. XX)} [qXq] 1[qXq][qXq] [qXq] 1[qXp][pXq] [qXp] 1[qXq][qXp] [qXp] 1[qXp][pXp] For δ(q.
X) = {(p. The productions of R also are to be renamed accordingly. Exercises: a. D. 0. Deterministic PDA: NPDA provides nondeterminism to PDA. A aAA.X) has at most one member for qQ. [pXp] to G and [pXq] to H.For δ(p. 1. X) = {(p. 1. XZ)} δ(q. [pZp] to D. Z) = {(q. CFG with productions: 1. A bBB  aS  a. PDA with transition function: δ(q. F. X) = {(p. X) = {(q. 0. . )} [pXp] 1 For δ(p.X) is not empty for some aΣ. 0. a Σ or a= and XΓ 2. [qZp] to B. G. B bA  a b. Convert to CFG. R. Z) = {(q.a. A. Z) = {(p. B. E. )} δ(p. F) is deterministic if and only if. [pZq] to C. Deterministic PDA’s (DPDA) are very useful for use in programming languages. q0.δ(q. X) = {(p. S aAS  bAB  aB. A aS  bS  a 2. S SS  (S)  3. the equivalent CFG can be defined by: G = ({S. S). For example Parsers used in YACC are DPDA’s. )} δ(p. H}. 1. 1. δ. . [qXq] to E [qXp] to F. X) = {(q. XX)} δ(q.1}.a. Σ. Z)} [pZq] 0[qZq] [pZp] 0[qZp] Renaming the variables [qZq] to A. Γ. then δ(q.X) must be empty 63 . )} II. C.If δ(q. Z0. 1. . Convert to PDA. X)} δ(q. {0. Definition: A PDA P= (Q. XX)} δ(p.
1/1 c. The transition diagram for the DPDA is given in figure 2.1/11 0. 0. To accept with final state: If L is a regular language. c Σ}. L=L(P) for some DPDA P. Accepting strings with number of a’s is more than number of b’s 3. but not by DPDA.0/00 1. Example: Construct DPDA which accepts the language L = {wcwR  w {a. The Context Free Languages could be recognized by nPDA.Z)} for all p.a)=p. Accepting the language of balanced parentheses. 64 . Accepting {0n1m n m} DPDA and Regular Languages: The class of languages DPDA accepts is in between regular languages and CFLs. but the DPDA used to simulate a regular language does not use the stack.0/ 10 q0 q1 . The two modes of acceptance are not same for DPDA. b}*.a. q Q such that δA(q.0/ 1. (Consider any type of parentheses) 2. Z0/1Z0 0. PDA surely includes a stack. Z0/ Z0 0.1/ c. then δP(q. NPDA can be constructed for accepting language of palindromes. The DPDA languages include all regular languages. The stack is inactive always.Z)={(p.1/ 01 1. If A is the FA for accepting the language L.DPDA is less powerful than nPDA. The class of language DPDA accept is in between than of Regular language and CFL.0/0 c. Z0/0Z0 1. Z0 / Z0 q2 Figure 2: DPDA L = {wcwR} Exercises: Construct DPDA for the following: 1.
So add a new production $ as a variable of G. then surely L has unambiguous CFG. y L. y L(WcWR). L={0*} could be accepted by DPDA with final state. 65 . But the language. For example S 0S01S1 is an unambiguous grammar corresponds to the language of palindromes. This is language is accepted by only nPDA. Definition of prefix property of L states that if x. N(P) L DPDA and Ambiguous grammar: DPDA is very important to design of programming languages because languages DPDA accept are unambiguous grammars. To convert L(P) to N(P) to have prefix property by adding an end marker $ to strings of L. So N(P) are properly included in CFL L. If L = L(P) for DPDA P. But all unambiguous grammars are not accepted by DPDA. x and y satisfy the prefix property. but not with empty stack. then L has unambiguous CFG. Then convert N(P) to CFG G’. because strings of this language do not satisfy the prefix property.To accept with empty stack: Every regular language is not N(P) for some DPDA P. A language L = N(P) for some DPDA P if and only if L has prefix property. ie. NonRegular language L=WcWR could be accepted by DPDA with empty stack. From G’ we have to construct G to accept L by getting rid of $ . or vice versa. because if you take any x. If L = N(P) for DPDA P. then x should not be a prefix of y.
A aA.Simplification of CFG The goal is to take an arbitrary Context Free Grammar G = (V. P1. S) be a grammar without useless symbols by 1. A CFG can be simplified by eliminating 1. A and B are variables 1. 2. then it is useless.Productions A . X is reachable. T. X is generating variable. P. Eliminating non generating symbols 2. T1. ie if X * w. If X is not useful. Useless symbols Those variables or terminals that do not appear in any derivation of a terminal string starting from Start variable. S) if there is S * X *w. where A is a variable 3. P. Omitting useless symbols from a grammar does not change the language generated Example: S aSb   A. Here B is a useless symbol Symbol X is useful if both the conditions given below are satisfied in that order itself. T. B bB. AaA  . . ie if S * X Theorem: Let G = (V. Eliminate useless symbols: Definition: Symbol X is useful for a grammar G = (V. T. S. P. Here A is a useless symbol SA. w*. S) and perform transformations on the grammar that preserve the language generated by the grammar but reach a specific format for the productions. Eliminating symbols that are non reachable 66 . S) be a CFG and assume that L(G) . 1. Unit production A B. then G1=(V1. where w* 2.
A}. S) Example: Eliminate the nongenerating symbols from S aS  A  C. A. S aS  A. the dependency graph is given below. If A and is already generating. S) Solution: B is a nongenerating symbol. B aa Eliminate symbols that are non reachable: Dependency Graph: For the production C xDy. After eliminating C gets. A a. Here A is still useless symbol. A a. C D 67 . so completely useless symbols are not eliminated. B}. If the order of eliminations is 1 followed by 2 gives S a. B aa. we get G1= ({S. Example: Consider a grammar with productions: S ABa. After eliminating B. {a}. {a}. S AB  a. If the order of eliminations is 2 followed by 1 gives S a. Every symbol of T is generating 2. A a}. A a. then A is generating Nongenerating symbols = V. A a. A a}.generating symbols. {S a.Elimination has to be performed only in the order of 1 followed by 2. Example: G= ({S. C aCb Solution: C is a nongenerating symbol. Eliminate nongenerating symbols: Generating symbols follow to one of the categories below: 1. Otherwise the grammar produced will not be completely useless symbols eliminated.
{S a. A}. G1= ({S}. A a}. Example: Eliminate nonreachable symbols from G= ({S. S) Example: Eliminate nonreachable symbols from S aS  A. A a. After eliminating B. B BC  AB. After eliminating A. S} P1 = {S CA. A a Example: Eliminate useless symbols from the grammar with productions S AB  CA. A a. {a}. {S a}. S) Solution: S A Draw the dependency graph as given above. B is nonreachable from S. B aa S A B Draw the dependency graph as given above. A a. C b} Step 2: Eliminate symbols that are non reachable S A C 68 . If there is no edge reaching a variable X from Start symbol S. A is nonreachable from S. {a}. then X is non reachable.Draw dependency graph for all productions. we get the grammar with productions S aS  A. C. C AB  b Step 1: Eliminate nongenerating symbols V1 = {A.
D E. B is nullable because of the production B . C abb. All Variables are reachable. E aC} P= {S aBa  BC. Eliminate . A is nullable because of the production A . 2. B bB  The procedure to find out an equivalent G with out productions 1. 69 .Draw the dependency graph as given above. S} P2 = {S CA. A Sb  bCC. A aA  . 3. In the grammar with productions S ABA. C b} Exercises: Eliminate useless symbols from the grammar 1. E d} P= {S aAa. A bBB. Find nullable variables 2. Example for a grammar G with . 3. If A * then A is a nullable variable. A bA.Productions: Most theorems and methods about grammars G assume L(G) does not contain . V2 = {A. S is nullable because both A and B are nullable. 4. C. B bB  . B bcc. C a. B AA} III. then we have to find out an equivalent G without productions.productions is S ABA. C aB} P= {S aS  AB. Add productions with nullable variables removed. A aA  . A aC  BCC. Remove productions and duplicates Step 1: Find set of nullable variables Nullable variables: Variables that can be replaced by null (). B ab. So the final variables and productions are same V1 and P1. P= {S aAa. A a. So if is not there in L(G).
A. minus any productions of the form A . B bB  . create all possible productions of the form A w’. production A } Repeat Ni Ni1U {A A in V.Productions G = ({S. B. Example: For the above example we get the final grammar with productions S ABA  BA  AA  AB  A  B A aA  a B bB  b Example: Find out the grammar without . So after removing nullable variables we get the productions S ABA  BA  AA  AB  A  B  A aA   a B bB   b Step 3: Remove productions and duplicates The desired grammar consists of the original productions together with the productions constructed in step 2. A .Algorithm to find nullable variables is given below: V: set of variables N0 {A  A in V. D}. S) 70 . A aA  . A α. Example: In the grammar with productions S ABA. B . {S aS  AB. {a}. where w’ is obtained from w by removing one or more occurrences of nullable variables. B and S are nullable variables. D b}. A. α in Ni1} until Ni = Ni1 Step 2: Add productions with nullable variables removed For each production of the form A w.
A aA  bB  . A aB . Y b  X S Xa. C aCD.productions and obtain G1 2. 3.Solution: Nullable variables = {S. D ddd Step 1: Eliminate . B. A. S) Exercise: Eliminate . follow the sequence given below: 1. B aA.productions from the grammar 1. D b}. W Z. D ddd Step 2: Eliminating useless symbols Step 2a: Eliminate nongenerating symbols 71 . Y bW. X Y . B} New Set of productions: S aS  a S AB  A  B Db G1= ({S. D}. 4. A aAS  a. Eliminate useless symbols from G1 and obtain G2 Example: Eliminate . X Zb. Z AB. 2. S a Xb  aYa.productions and useless symbols from the grammar S a aABC.productions and useless symbols. C aCD. B Ba  Bb S ASB  . A aB. {a}. Eliminate . {S aS  a  AB  A  B. X aX  bX  S XY.productions Nullable ={A} P1={S a aA  B  C. B SbS  A bb But if you have to get a grammar without . B aAa.
D ddd} Step 2b: Eliminate non reachable symbols S A B D C is nonreachable. B aAa. B.Generating ={D. S} P2={S a  aA B. A aB. A. B aAa} 72 . A aB. eliminating C gets P3= S a aAB.
Now add productions from derivable. The algorithm for eliminating unit productions from the set of production P is given below: 1. Unit productions could complicate certain proofs and they also introduce extra steps into derivations that technically need not be there. S B. Add all non unit productions of P to P1 2. S ABA  BA  AA  AB  A  B  aA  a  bB  b A aA  a B bB  b Remove unit productions from above productions to get S ABA  BA  AA  AB  aA  a  bB  b A aA  a B bB  b Example: Eliminate unit productions from S Aa  B. 3. where B is a nonunit production in P. B bBb Solution: The unit productions are S A. A aA  a. A a  bc  B. add to P1 all productions A . Delete all the unit productions Example: Eliminate unit productions from S ABA  BA  AA  AB  A  B. For each unit production A B. where A and B are variables. So A and B are derivable. B A  bb Solution: 73 .I. Definition: Eliminate unit productions: Unit production is of form A B.
follow the sequence given below: 1. C cCD. D ddd} After eliminate symbols that are non reachable 74 . D ddd Step 1: Eliminate productions Nullable = {A} P1 = {S a  aA  B  C. C cCD. B aA. Eliminate . useless symbols and unit productions. C cCD. A B and B A.Unit productions are S B. A aB. B aA  a.productions from G and obtain G1 2. B aA  a.productions. B aA  a. Eliminate useless symbols from G2and obtain G3 Example: Eliminate useless symbols. productions and unit productions from the grammar with productions: S a  aA  B C. So Derivable variables are B and C. Eliminate unit productions from G1 and obtain G2 3. A aB  . P2 = {S a  aA  cCD. A aB. So A and B are derivable. D ddd} Step 2: Eliminate unit productions Unit productions are S B and S C. A aB. S Aa  bb  a  bc A a  bc  bb B bb  a  bc Simplified Grammar: If you have to get a grammar without . Add productions from derivable and eliminate unit productions to get. D ddd} Step 3: Eliminate useless symbols After eliminate nongenerating symbol C we get P3 = {S a  aA.
A aB. … . {S a  aA. C2. Cb b 75 . then add a new variable Cai to V1 and a new production Cai ai to P1. has a grammar in CNF with productions of the form: 1. B}. Elimination of terminals on RHS of a production a) Add all productions of the form A BC or A a to P1 b) Consider a production A X1X2…Xn with some terminals of RHS. productions and unit productions from the grammar 2. B aA  a} So the equivalent grammar G1 = ({S. where A V and a T Algorithm to produce a grammar in CNF: 1. D d Solution: Step1: Simplify the grammar Grammar is already simplified. Introduce new productions A X1C1. A aB  bAB. B.S A B D P4 = {S a  aA. If Xi is a terminal say ai. where A. {a}. Chomsky Normal Form (CNF) Every nonempty CFL without . Eliminate useless symbols. C1 X2C2. A BC. S) II. A a. Cn2 to V1 Example: Convert to CNF: S aAD. where n 3 and all Xi‘s are variables. Cn2 Xn1Xn to P1 and C1. … . A aB. Step2a: Elimination of terminals on RHS Change S aAD to S CaAD. Ca a A aB to A CaB A bAB to A CbAB. B b. Replace Xi in A production of P by Cai c) Consider A X1X2…Xn. A. C V 2. B aA  a}.
A CaB CbC2. Ca. C2}. A aAS  aA  a. C2 AS. S) Example: Convert the grammar with following productions to CNF: P={S ASB  . A CaC2  CaA  a. Ca a. B SC3  SCb  CbS  b  CbCb  CaC2  CaA  a. Cbb. Cb b. S CaC1. S aSa  bSb  a  b  aa  bb 76 . Caa. A aAS  a. B. B SbS  Sb  bS  b  A  bb} Step 1b: Eliminate unit productions: B A P2= {S ASBAB. Cb. C2 AB Grammar converted to CNF is given below: G1=({S. b}. B SbS  A  bb} Solution: Step1: Simplify the grammar Step 1a: Eliminate productions: Consists of S Eliminating productions from P to get: P1={S ASBAB. C1 AD. C1 SB. C2AB}. C1. C3 CbS} Exercises: Convert the following grammar with productions to CNF: 1. A aASaAa. {a. D. C1 AD A CbAB to A CbC2. A.Step2b: Reduce RHS with 2 variables Change S CaAD to S CaC1. B SbS  Sb  bS  b  bb  aAS  aA  a} Step 1c: Eliminate useless symbols: no useless symbols Step2: Convert to CNF P3={S AC1  AB.
Example: Transform into Greibach normal form the grammar with productions S 0S1  01 Solution: S 0SA  0A. C S S aAa  bBb . Greibach Normal Form (GNF): Every nonempty language without is L(G) for some grammar G with productions are of the form A a. A bAA  aS  a. The remainder of the computation uses the input symbol and the stack top to determine the appropriate transition. 4. B aBB  bS  b S Aba. B SA. A aab. A C. 5. A Ca. D A  B  ab III. A 1 Usage of GNF: Construction of PDA from a GNF grammar can be made more meaningful with GNF. C CDE . Conversion to GNF is a complex process. An S rule of the form S aA1A2…An generates a transition that processes the terminal a. where a T and is a string of zero or more variables. 3. pushes the variables A1A2…An on the stack and enters q1. Converting CFG in GNF to PDA gets a PDA without rules. S bA  aB. Assume that the PDA has two states: start state q0 and accepting state q1.2. 77 . B AC S 0A0 1B1  BB. B C  b.
2 symbols at 1. That is. consider a CFL L={anbn  n 1}. The extended parse tree for the string a4b4 is given in figure 2. Equivalent CNF grammar is having productions S AC  AB. as often as we like. C SB. Both leftmost derivation and rightmost derivation have same parse tree because the grammar is unambiguous. The pumping lemma for CFL’s states that there are always two short substrings close together that can be repeated. 1 symbols at level 0. tree must be having at least depth of n and level of at least n+1. The parse tree for the string a4b4 is given in figure 1. To have 2n symbols at bottom level. both the same number of times. For example. Pumping Lemma Theorem: 78 . Number of symbols at each level is at most twice of previous level. A a. B b. Figure 1: Parse tree for a4b4 Figure 2: Extended Parse tree for a4b4 Extend the tree by duplicating the terminals generated at each level on all lower levels.The Pumping Lemma for CFL The pumping lemma for regular languages states that every sufficiently long string in a regular language contains a short substring that can be pumped. 4 symbols at 2 …2i symbols at level i. inserting as many copies of the substring as we like always yields a string in the regular language.
Figure 4: sub. By pigeonhole principle. Say X has 2 occurrences. at least one of the strings we pump must not be empty). and the parse tree is shown in figure 3. 3. Example parse tree: For the above example S has repeated occurrences. Figure 3: Parse tree for a4b4 with repeated occurrences of S circled. consider the first pair of same variable along the path. then we can write z = uvwxy such that 1. Reading from bottom to top. 2. y=bb. vwx k (that is. Suppose z L(G) and z k. Then there exists a constant k 0 such that if z is any string in L such that z k. w = ab is the string generated by lower occurrence of S and vwx = aabb is the string generated by upper occurrence of S. uviwxiy is in L. Proof: The parse tree for a grammar G in CNF will be a binary tree. v=a. Let k = 2n+1. w=ab. the middle portion is not too long). Break z into uvwxy such that w is the string of terminals generated at the lower occurrence of X and vwx is the string generated by upper occurrence of X. So here u=aa. x=b. The longest path in the parse tree is at least n+1. Any parse tree for z must be of depth at least n+1. where n is the number of variables of G. For all i 0. vx (since v and x are the pieces to be “pumped”.Let L be a CFL. some variables occur more than once along the path. so this path must contain at least n+1 occurrences of the variables.trees 79 .
These parse trees are shown in figure 4. 4. 2. 3. To show that a language L is not a CFL. Figure 5: Parse tree for uv2wx2y L Figure 6: Parse tree for uwy L To get uwy L. cut out t and replace it with copy of T. Choose an “appropriate” string z in L Express z = uvwxy following rules of pumping lemma Show that uvkwxky is not in L. cut T out of the original tree and replace it with t to get a parse tree of uv0wx0y = uwy as shown in figure 6. 5. Pumping Lemma game: 1. Since vwx n then vwx can either consists of 80 . assume L is context free. 6. To get uv2wx2y L.Let T be the subtree rooted at upper occurrence of S and t be subtree rooted at lower occurrence of S. The parse tree for uv2wx2y L is given in figure 5. Choose an appropriate z = anbncn = uvwxy. Cutting out t and replacing it with copy of T as many times to get a valid parse tree for uviwxiy for i 1. for some k The above contradicts the Pumping Lemma Our assumption that L is context free is wrong Example: Show that L = {aibici  i 1} is not CFL Solution: Assume L is CFL.
then uv2wx2y will be a2b3c2 L If you consider any of the above 3 cases. v = b. According to pumping lemma. v = a. Some a’s and some b’s 3. But Pumping Lemma says uv2wx2y L. Both t’s are not same.1. uv2wx2y will be a4b2c2L Case 2: vwx consists of some a’s and some b’s If z = a2b2c2 and u = a. y = c. x = b. then uv2wx2y will be a3b3c2 L Case 3: vwx consists of some b’s and some c’s If z = a2b2c2 and u = a2b. n t 3n/2. It is sufficient to show that L1= {0m1n0m1n  m. t= 81 . This language we prove by taking the case of i = 0. So if vwx n. w = c. If uwy is some repeating string tt. v = a. Example: z = 03130313. All a’s or all b’s or all c’s 2. y = bc2. Example: Show that L = {ww w {0. uv2wx2y will not be having an equal number of a’s. 1}*} is not CFL Solution: Assume L is CFL. x = . Pick any z = 0n1n0n1n = uvwxy. so first t = 0130 and second 0213. b’s and c’s. w = . Then uwy begins with 0nk1n uwy = 4nk. z is having a length of 4n. uwy L. So second t is not a repetition of first t. t does end in 0 but tt ends with 1. satisfying the conditions vwx n and vx . w = . where n is pumping lemma constant. then uwy 3n. is a CFL. So L is not contextfree. x = a and y = b2c2 then. Can’t contradict the pumping lemma! Our original assumption must be wrong.n 0}. Suppose vwx is within first n 0’s: let vx consists of k 0’s. then t =2nk/2. where t is repeating. Then uwy will be some string in the form of tt. vx = 02 then uwy = tt = 0130313. If so. Some b’s and some c’s Case 1: vwx consists of all a’s If z = a2b2c2 and u = . in the pumping lemma satisfying the condition uviwxiy for i 0.
So L is not a CFL. which is not CFL. If vx has at least one 1. then uwy is not in the form tt. we get uwy. Very similar explanations could be given for the cases of vwx consists of first block of 1’s and vwx consists of 1st block of 1’s and 2nd block of 0’s. If we consider any combinations of above cases. But first t and second t are not the same string. then t is at least 3n/2 and first t ends with a 0. Exercises: Using pumping lemma for CFL prove that below languages are not context free 1. {0p  p is a prime} 2. Case 2: vwx consists of 2 adjacent symbols say 1 & 2 Then uwy is missing some 1’s or 2’s and uwy is not in L. Example: Show that L={0i1j2i3j  i 1. vwx can consist of a substring of one of the symbols or straddles of two adjacent symbols. So uwy is not in L and L is not context free. {anbnci  i n} 82 . not a 1. Pick z = uvwxy = 0n1n2n3n where vwx n and vx . j 1} is not CFL Solution: Assume L is CFL. In all cases uwy is expected to be in the form of tt.Suppose vwx consists of 1st block of 0’s and first block of 1’s: vx consists of only 0’s if x= . Case 1: vwx consists of a substring of one of the symbols Then uwy has n of 3 different symbols and fewer than n of 4th symbol. This contradicts the assumption. Then uwy is not in L.
generated by the grammar S abA.bb} is a substitution on alphabet ={0. closure (star). I. P. T. abc}. and setdifference. concatenation. union. CFL’s are not closed under intersection (but the intersection of a CFL and a regular language is always a CFL). Rename second and third S’s to S0 and S1. with each terminal a is replaced by Sa everywhere a occurs. Pa. complementation. homomorphism and inverse homomorphism.Closure Properties of CFL Many operations on Context Free Languages (CFL) guarantee to produce CFL. Rename second A to B. Closure properties consider operations on CFL that are guaranteed to produce a CFL. s(0) = {anbm  m n}. Example: L = {0n1n n 1}. Proof: Let G = (V. generated by the grammar S 0S1  01. s(1) = {ab. generated by the grammar S aSb  A. Example: S(0) = {anbn n 1}. S) is a grammar for s(L) where V = V Va T= union of Ta’s all for a P consists of o All productions in any Pa for a o The productions of P. Theorem: If a substitution s assigns a CFL to every symbol in the alphabet of a CFL L. Substitution: By substitution operation. Then G= (V. each symbol in the strings of one language is replaced by an entire CFL language. S) be grammar for the CFL L. then s(L) is a CFL. A aA  ab. S(1)={aa. . The CFL’s are closed under substitution. A c . Let G a = (Va. respectively. Ta. A few do not produce CFL. 1}. P. Resulting grammars are: 83 . reversal. Sa) be the grammar corresponding to each terminal a and V Va = .
P. S (V1 V2) and P = {P1 P2 {S S1  S2 }} Example: L1 = {anbn  n 0}.S 0S1  01 S0 aS0b  A. S V1 V2 and P = P1 P2 {S S1S2} 84 . b}. L2 = {bnan  n 0}. s(a)=L1 and s(b)=L2. {S S1  S2. B c  In the first grammar replace 0 by S0 and 1 by S1. Closure under union of CFL’s L1 and L2: Use L={a. Then s(L)=L1L2 How to get grammar for L1L2? Add new start symbol and rule S S1S2 The grammar for L1L2 is G = (V. The resulted grammar after substitution is: S S0SS1  S0S1 S0 aS0b  A. S1. How to get grammar for L1 L2 ? Add new start symbol S and rules S S1  S2 The grammar for L1 L2 is G = (V. S2}. S1 aS1b  . S). A aA  ab S1abB. T. S2 bS2a}. {a. Application of substitution: a. Then s(L)= L1 L2. s(a)=L1 and s(b)=L2. A aA  ab S1 abB. B c  II. Closure under concatenation of CFL’s L1 and L2: Let L={ab}. b}. Their corresponding grammars are G1: S1 aS1b  . P. T. b. G2 : S2 bS2a  The grammar for L1 L2 is G = ({S. S) where V = V1 V2 {S}. S) where V = {V1 V2 S}.
Closure under Kleene’s star (closure * and positive closure +) of CFL’s L1: Let L = {a}* (or L = {a}+) and s(a) = L1. Let s be a substitution that replaces every a . The grammar for (L1)* is G = (V.e. P= P1 {S SS1  } d. S). G2 : S2 bS2a  The grammar for L1L2 is G = ({S. III. S). c. L2 = {bnan  n 0} then L1L2 = {anb{n+m}am  n. Example: L1 = {anbn  n 0} (L1)* = {a{n1}b{n1} . Then h(L) = s(L). b}. IV.. ie h(L) ={h(a1)…h(ak)  k 0} where h(ai) is a homomorphism for every ai . S1 aS1b  . to substitute each production A by AR.. ie s(a) = {h(a)}. P. {S S1S2. i. S2}. S1. Closure under homomorphism of CFL Li for every ai: Suppose L is a CFL over alphabet and h is a homomorphism on . S V1. by h(a). T. a{nk}b{nk}  k 0 and ni 0 for all i} L2 = {a{n2}  n 1}. (L2)* = a* How to get grammar for (L1)*: Add new start symbol S and rules S SS1  . Intersection: The CFL’s are not closed under intersection 85 . m 0} Their corresponding grammars are G1: S1 aS1b  . so LR is a CFL. Then s(L) = L1* (or s(L) = L1+).. It is enough to reverse each production of a CFL for L. Closure under Reversal: L is a CFL.Example: L1 = {anbn  n 0}. S2 bS2a}. where V = V1 {S}. {a.
s = A(p. p). . . ) are possible if and only if (qp. then L R is a CFL. ie w is in L R. we make the same move in PDA P and also we carry along the state of DFA A in a second component of P. qA) F = (FPX FA) is in the form ((q. . To get L R. But L = L1 L2. Corresponding grammars for L1: SAB. B2B  2 and corresponding grammars for L2: S AB. qA. . a. a. Let A = (QA. X) That is for each move of PDA P.) moves and p = *(qA. w. B1B2  12. s). . Z) *P ((q. The moves ((qp. Z0. FA) for DFA to accept the Regular Language R. i 1} is also a CFL. FA AND PDA Accept/ Reject Stack Proof: Figure 1: PDA for L R P = (QP. . Z0. g) such that 1. qo. But L1 = {0n1n2i  n 1. (r. A0A  0. thus intersection of CFL’s is not CFL a. . we have to run a Finite Automata in parallel with a push down automata as shown in figure 1. b. p). P. However. Z) *P (q. FP) be PDA to accept L by final state. A0A1  01. X) = ((r. . w. qA). Intersection of CFL and Regular Language: Theorem: If L is CFL and R is a regular language. qP. A. L = L1 L2 .Example: The language L = {0n1n2n  n 1} is not contextfree. w) transitions are possible. P accepts a string w if and only if both P and A accept w. i 1} is a CFL and L2 = {0i1n2n  n 1. CFL and RL properties: 86 . g) is in P(q. a) 2. Construct PDA P = (Q. F) where Q = (Qp X QA) qo = (qp.
L2 is not necessarily a CFL. is the set of all strings w such that h(w) L. called an inverse homomorphism. But CFL’s are not closed under complementation. So RC is also regular. Inverse Homomorphism: Recall that if h is a homomorphism. V. L1C L2C is a CFL. and a regular language R. But (L1C L2C)C = L1 L2. Proof: Let L1 = * . Closure of CFL’s under setdifference with a regular language. 2. CFL is not closed under complementation LC is not necessarily a CFL Proof: Assume that CFLs were closed under complement.R is a CFL. ie L . L1. Since CFLs are closed under union. which we just showed isn’t necessarily a CFL. So CFLs are not closed under setdifference. If CFLs were closed under set difference. So our assumption is false. 1. and L is any language.L = LC.L = LC would always be a CFL. ie if L is a CFL then LC is a CFL. Theorem: If L is a CFL and h is a homomorphism. then h1(L) is a CFL 87 . We have already proved the closure of intersection of a CFL and a regular language. The CFL’s are closed under inverse homomorphism.L. ie L1 . But * . Proof: R is regular and regular language is closed under complement.R = L RC. then * . Contradiction! . 3. By our assumption (L1C L2C)C is a CFL. * is regular and is also CFL. then h 1(L). So CFL is closed under set difference with a Regular language. CFL is not closed under complementation. We know that L .Theorem: The following are true about CFL’s L. CFLs are not closed under setdifference. and L2.
88 .
). ). Once we accept the relationship between P and P. Γ. a. bx). where Q is the set of pairs (q. T. because of the way the accepting states of P are defined. b. δ. )} where b T or b = then δ ((q. where q is an accepting state of P. . Let PDA P = (Q. ). Z0.X)} o If δ(q. (F x )) to accept h1(L). We construct a new PDA P = (Q. δ. Suppose h applies to symbols of alphabet Σ and produces strings in T*.Buffer a Input h h(a) PDA Accept/ Reject Stack Figure 2: PDA to simulate inverse homomorphism We can prove closure of CFL under inverse homomorphism by designing a new PDA as shown in figure 2. Symbols of h(a) are used one at a time and fed to PDA being simulated. x). Z0. Σ. q0. ) The accepting state of P is (q. )} The start state of P is (q0. X) = {(p. h(a)). x) such that o q is a state in Q o x is a suffix of some string h(a) for some input string a in Σ δ is defined by o δ ((q. X) = {((p. F) that accept CFL L by final state. X) = {((q. h(a) is placed in a buffer. Only when the buffer is empty does the PDA read another of its input symbol and apply homomorphism to it. Thus L(P)=h1(L(P)) 89 . (q0. P accepts h(w) if and only if P accepts w. Γ. After input a is read.
e. Ex: To recognize string that is a multiple of 4 1 0 0 0 S2 0 S1 S0 90 1 . language PDA Context free language \ Regular language DFA Regular language A language is called a regular language if some finite automaton recognizes it.Language Hierarchy and History of Turing Machine A Hierarchy of Formal Languages Turing machines r.
e.0 ) (S 3 .0 .1 ) (S 3 .L) ( S 3. language) A language is called r.e language if some Turing machine recognizes it.But can regular language recognize strings of the form 0n1n ? No Context Free Language A language is called Context free language iff some pushdown automaton recognizes it.L) ( S 3 .Y)(SA.R) SA : Accepting state 91 .1.Y.Y. Ex: To recognize a string of the form 0n1n S 0S1/ Limitation again Can context free language recognize strings of the form 0n1n2n ? No Recursively Enumerable Language (r.R) ( S 2 . Ex: To recognize strings of the form 0n1n2n ( S 0 .0 .0 ) (S1.R) ( S 0 .1 ) (S 2 .1.Y.R) ( S 1.0 ) (S1.R) ( S 1.X.L) ( S 3 .1 ) (S 2 .X.Z.R) ( S 2 .X)(S 0 .L) ( S 3 .Y)(S 3.2) (S 3.
(0  1)* Regular Expn.Formal Machines Overview Regex operators: * (Kleene *)  (choice) DFA Regex example: . (concat) Regular Lang. NFA with moves NFA 2way DFA DFA 92 .
and a criteria for membership. addressed Hilbert’s Entscheidungsproblem using a different approach.PDA CFL example: 0n1n CFL CFG But there are strings that cannot be recognized by PDAs. and a universe of elements U. is it possible to formulate a general procedureto decide whether a given element of U is a member of X? Hilbert’s grand ideas were watered down by the incompleteness theorem proposed by the Austrian Kurt Gödel. He proposed two kinds of mathematical 93 . His theorem states that in any mathematical system. there exist certain obviously true assertions that cannot be proved to be true by the system. In 1936 the British cryptologist Alan Turing. Given a set X. For example: anbncn Multistack PDA PDA with queues PDA Some Historical Notes At the turn of the 20th century the German mathematician David Hilbert proposed the Entscheidungsproblem.
machines called the amachine and the cmachine respectively, and showed the existence of some problems where membership cannot be determined. But Turing’s amachine became famous for something else other than its original intentions. The amachine was found to be more expressive than CFGs. It can recognize strings that cannot be modeled by CFLs like anbncn. The amachines came to be more popularly known as Turing Machines The Princeton mathematician Alonzo Church recognized the power of amachines. He invited Turing to Princeton to compare Turing Machines with his own lcalculus. Church and Turing proved the equivalence of Turing Machines and lcalculus, and showed that they represent algorithmic computation. This is called the ChurchTuring thesis.
Turing Machines
Definition: A Turing Machine (TM) is an abstract, mathematical model that describes what can and cannot be computed. A Turing Machine consists of a tape of infinite length, on which input is provided as a finite sequence of symbols. A head reads the input tape. The Turing Machine starts at “start state” S0. On reading an input symbol it optionally replaces it with another symbol, changes its internal state and moves one cell to the right or left.
Notation for the Turing Machine : TM = <S, T, S0, , H> where, S is a set of TM states T is a set of tape symbols S0 is the start state HS is a set of halting states : S x T S x T x {L,R} is the transition function {L,R} is direction in which the head moves L : Left R: Right
94
input symbols on infinite length tape
1 0 1 0 1 1 1 1 1 1 0
head
The Turing machine model uses an infinite tape as its unlimited memory. (This is important because it helps to show that there are tasks that these machines cannot perform, even though unlimited memory and unlimited time is given.) The input symbols occupy some of the tape’s cells, and other cells contain blank symbols.
Some of the characteristics of a Turing machine are: 1. The symbols can be both read from the tape and written on it. 2. The TM head can move in either directions – Left or Right. 3. The tape is of infinite length 4. The special states, Halting states and Accepting states, take immediate effect.
Solved examples: TM Example 1: Turing Machine U+1: Given a string of 1s on a tape (followed by an infinite number of 0s), add one more 1 at the end of the string. Input : #111100000000……. Output : #1111100000000………. Initially the TM is in Start state S0. Move right as long as the input symbol is 1. When a 0 is encountered, replace it with 1 and halt. Transitions: (S0, 1) (S0, 1, R) (S0, 0) ( h , 1, STOP) TM Example 2 : 95
TM: XY Given two unary numbers x and y, compute xy using a TM. For purposes of simplicity we shall be using multiple tape symbols. Ex: 5 (11111) – 3 (111) = 2 (11) #11111b1110000….. #___11b___000… a) Stamp out the first 1 of x and seek the first 1 of y. (S0, 1) (S0, b) (S1, 1) (S1, b) (S1, _, R) (h, b, STOP) (S1, 1, R) (S2, b, R)
b) Once the first 1 of y is reached, stamp it out. If instead the input ends, then y has finished. But in x, we have stamped out one extra 1, which we should replace. So, go to some state s5 which can handle this. (S2, 1) (S2,_) (S2, 0) (S3, _, L) (S2, _, R) (S5, 0, L)
c) State s3 is when corresponding 1s from both x and y have been stamped out. Now go back to x to find the next 1 to stamp. While searching for the next 1 from x, if we reach the head of tape, then stop. (S3, _) (S3,b) (S4, 1) (S4, _) (S4, #) (S3, _, L) (S4, b, L) (S4, 1, L) (S0, _, R) (h, #, STOP)
d) State s5 is when y ended while we were looking for a 1 to stamp. This means we have stamped out one extra 1 in x. So, go back to x, and replace the blank character with 1 and stop the process. (S5, _) (S5,b) (S6, 1) (S6, _) (S5, _, L) (S6, b, L) (S6, 1, L) (h, 1, STOP)
96
R) S3 = Seeking X.Y._ .Y)(S 4 .STOP) 97 ._ ) (SA. then seek the first 1 and stamp it with Y.Z)(S 4 .Z.X.Y.1 ) (S 2 .R) ( S 1.R) ( S 2 .L) ( S 3.0 . stamp it with X S1 = Seeking 1. stamp it with Y S2 = Seeking 2.0 ) (S1.Y. stamp it with Z Step 2: Move left until an X is reached.0 ) (S1. seeking 0. to repeat the process.R) ( S 4 .0 ) (S 3.1 ) (S 3.2) (S 3 .Y)(S 4 .0.1 ) (S 2 .X)(S 0 . and then seek the first 2 and stamp it with Z and then move left.L) S0 = Start State.R) ( S 4 .R) ( S 1.Design of Turing Machines and Universal Turing Machine Solved examples: TM Example 1: Design a Turing Machine to recognize 0n1n2n ex: #000111222_ _ _ _ _…….1.Y)(S 3.R) ( S 4 . ( S 3. then move one step right. then the accepting state SA is reached.X. ( S 0 . Step 1: Stamp the first 0 with X. Step 3: Move right until the end of the input denoted by blank( _ ) is reached passing through X Y Z s only.Z.R) ( S 2 .1.L) ( S 3. ( S 0 .Y.L) ( S 3.
1) (S 6 .1 ) (h.2 ) (h.STOP) ( S 1._ . ( S 3 .2 .1) (S 6 .1.R) ( S 1.0) (S 6 . ( S 0 .R) ( S 0 ._ .L) ( S 5 .STOP) 1 ( S 4 .2 ) (h.STOP) ( S 4 .2 .L) ( S 4 .2 ) (h._ . then move left until a blank is reached to start the process again.1 ) (S 2 ._ ) (S 5 ._) (S 0 .0) (S 4 .L) ( S 3.2 .1) (S 4 ._ ) (h.0 ) (h.1 ) (h. ._ .L) ( S 6 .R) 98 .STOP) ( S 2 ._ .0 ) (S1._ ) (SA.L) ( S 6 .STOP) TM Example 2 : Design a Turing machine to accept a Palindrome ex: #1011101_ _ _ _ _……. ( S 4 . then seek the last character by moving till a _ is reached.L) ( S 4 .S4 = Seeking blank These are the transitions that result in halting states.STOP) ( S 2 .0.1._) (S 0 ._ ._ ) (S 3.STOP) ( S 0 .1.L) ( S 4 ._ .STOP) 1 ( S 0 ._ .0) (S 4 .0._ .0 .1 ) (h. .STOP) Step 2: If the last character is 0/1 accordingly. If the last character is not 0/1 (as required) then halt the process immediately._ .L) ( S 6 . Step 1: Stamp the first character (0/1) with _.R) ( S 5 .
Design a TM to recognize a string of the form a b ..STOP) The sequence of events for the above given input are as follows: #s010101_ _ _ #_s20101_ _ _ #_0s2101_ _ _ .. The associations between the modules are shown in the following figure: 99 . the problem of designing Turing machine to recognize the language 0n1n2n can be divided into modules such as 0stamper. ( S 3._) (SA.._ .Step 3 : If a blank ( _ ) is reached when seeking next pair of characters to match or when seeking a matching character. then accepting state is reached. #_ _ _ _ s5 _ _ _ _ _ _ #_ _ _ _ sA _ _ _ _ _ _ Exercises: n 2n 1.STOP) ( S 0 . The main problem can be divided into sequence of modules. 2. 1stamper._ . Inside each module._ . #_0101s5_ _ _ #_010s6_ _ _ _ #_s60101_ _ _ #_s00101_ _ _ . Modularization of TMs Designing complex TM s can be done using modular approach. For example... 2seeker and 2stamper. 0seeker. there could be several state transitions.. Design a TM to recognize a string of 0s and 1s such that the number of 0s is not twice as that of 1s. 1seeker._) (SA._) (SA.STOP) ( S 5 .
H={b.1} : (a.0) (b. (c.R) .c. where $ is delimiter This spec along with the actual input data would be the input to the UTM.d} T={0. A TM spec could be as follows: TM = (S. S0=a.R) and so on then TM spec: $abcd$a$bd$01$a0b1Ra1c1Rc0d0R…….1.T.b. S={a. 100 .d) Suppose.1.1) (c.H.R) .0) (d.S0. (a. This can be encoded in binary by assigning numbers to each of the characters appearing in the TM spec.0.TM: 0n1n2n 0Stamper 1Seeker 1Stamper 2Seeker 2Stamper 0Seeker Universal Turing Machine A Universal Turing Machine UTM takes an encoding of a TM and the input data as its input in its tape and behaves as that TM on the input data.d}.
Sequence of actions in UTM: Initially UTM is in the start state S0. where necessary.0001. Make changes. Then repeat the steps with next input.0100.0011. Go back and find which transition to apply.The encoding can be as follows: $ : 0000 0 : 0101 a : 0001 1 : 0110 b : 0010 L : 0111 c : 0011 R : 1000 d : 0100 So the TM spec given in previous slide can be encoded as: 0000. the sequence goes through the cycle: Load DecodeExecuteStore 101 .0100 …… Hence TM spec can be regarded just as a number.0010.0000.0000. Load the input which is TM spec. Then store the changes.0010. Hence.0001.
Move back left without changing the tape 102 .Extensions to Turing Machines Proving Equivalence For any two machines M1 from class C1 and M2 from class C2: M2 is said to be at least as expressive as M1 if L(M2) = L(M1) or if M2 can simulate M1. 01. Equivalence of CTMs and TMs A CTM is simply a TM with a complex alphabet.. T = {a. d} T’ = {00. STM = <S. 10. do the following: 1. S} Equivalence of STMs and TMs STM = TM: Just don’t use the S option… TM = STM: For L and R moves of a given STM build a TM that moves correspondingly L or R… TM = STM: For S moves of the STM. 11} Turing Machines with Stay Option Turing Machines with stay option has a third option for movement of the TM head: left. R. b. Composite Tape TMs Track 0 0 1 1 0 1 0 1 0 0 … 0 0 1 1 1 1 1 1 0 … 1 Track A composite tape consists of many tracks which can be read or written simultaneously. A composite tape TM (CTM) contains more than one tracks in its tape. right or stay. 2. T. s0. c. .Move right. M1 is said to be at least as expressive as M2 if L(M1) = L(M2) or if M1 can simulate M2. H> : S x T à S x T x {L.
the tape is infinite on both sides.*. R) (s’’. b.STM: (s.(s’. Equivalence of 2TMs and TMs 2TM = TM: Just don’t use the left part of the tape… TM = 2TM: Simulate a 2way infinite tape on a oneway infinite tape… … 6 5 4 3 2 1 0 1 2 3 4 5 6 … 0 –1 1 –2 2 –3 3 –4 4 –5 5 … Multitape Turing Machines A multitape TM (MTM) utilizes many tapes.a) .a) .3.S) TM: (s.L) 2way Infinite Turing Machine In a 2way infinite TM (2TM).b. There is no # that delimits the left end of the tape. 103 .(s’’.(s’.*) .
Multidimensional TMs Multidimensional TMs (MDTMs) use a multidimensional space instead of a single dimensional tape..Equivalence of MTMs and TMs MTM = TM: Use just the first tape… TM = MTM: Reduction of multiple tapes to a single tape. A B C 0 1 2 3 4 5 6 7 … 0 1 2 3 4 5 6 7 … 0 1 2 3 4 5 6 7 … TM A0 B0 C0 A1 B1 C1 A2 B2 C2 A3 B3 . A single tape TM that is equivalent can be constructed by reducing m tapes to a single tape. Consider an MTM having m tapes. TM 104 .
R)} Equivalence of NTMs and TMs A “concurrent” view of an NTM: (s2.a.R} Ex: (s2.R)} è at (s2.a) à {(s3.b.a) à (s3. H> where : S x T à2SxTx{L.R) 105 . .Equivalence of MDTMs to TMs Reducing a multidimensional space to a single dimensional tape.a) à (s4. 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 Nondeterministic TM A nondeterministic TM (NTM) is defined as: NTM = <S.a).a.L) (s2.L) (s4. T.a) à {(s3.a.b.L) (s4. two TMs are spawned: (s2.b. s0.
L) (s4.a.R)} bccaaabccacb s2 bccbaabccacb s3 bccaaabccacb s4 106 .a) à {(s3.b.Simulating an NTM with an MDTM Consider an MDTM to simulate an NTM: (s2.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue listening from where you left off, or restart the preview.