Professional Documents
Culture Documents
Professor A G Constantinides© 1
AGC
DSP Signal Spaces
In order to compare two signals we need to define a
measure of “distance” known as a metric.
A metric is a function d ( x, y ) that produces a
scalar value from two signals such that
1)
d ( x, y ) d ( y , x )
2) d ( y, x) 0
3) d ( x, x ) 0
4) d ( x, z ) d ( x, y ) d ( y , z )
Professor A G Constantinides© 2
AGC
DSP Signal Spaces
There are many metrics that can be used.
Examples:
1) If we have a set of finite numbers representing
signals then a metric may be
n
d1 ( x, y ) xi yi
i 1
Professor A G Constantinides© 3
AGC
DSP Signal Spaces
2) Another metric for the same set is
n 1/ 2
d 2 ( x, y ) xi yi
2
i 1
This is called the l2 metric or the Euclidean
distance
3) Yet another form is the lp metric
n 1/ p
d p ( x, y ) xi yi
p
i 1
Professor A G Constantinides© 4
AGC
DSP Signal Spaces
4) As the integer p the last form becomes
d ( x, y ) max xi yi
i 1, 2 ,..., n
Professor A G Constantinides© 5
AGC
DSP Signal Spaces
When the set of vectors which we use is defined
along with an appropriate metric then we say we
have a metric space.
Professor A G Constantinides© 6
AGC
DSP Vector Spaces
We think of a vector x as an assembly of elements
xarranged
i i 1as
,2,3,4,...
x1
x
x 2
.
x
n
The length may be in certain cases infinite
Professor A G Constantinides© 7
AGC
DSP Vector Spaces
A linear vector space S over a set of scalars R is
formed when a collection of vectors is taken together
with an addition operation and a scalar multiplication,
and satisfies the following:
1) S is a group under addition ie the following are
satisfied
a) for any x and y in S then x y is also in S
b) there is a zero identity element ie
c) for every element there is another such that their sum is
zero x0x
d) addition is associative ie
(x y ) z x (y z )
Professor A G Constantinides© 8
AGC
DSP Vector Spaces
2) For any pair of scalars a, b and
R x, y S
ax S
a (bx) (ab)x
(a b)x ax bx
a (x y ) ax ay
3) There is a unit element in the set of scalars R such
that
1x x
(The set of scalars is usually taken as the set of real
numbers)
Professor A G Constantinides© 9
AGC
DSP Linear Combination
Often we think of a signal as being composed of
other simpler (or more convenient to deal with)
signals. The simplest composition is a linear
combination of the form
m
x[n] ci pi [n]
i 1
Professor A G Constantinides© 10
AGC
DSP Vector space …
Professor A G Constantinides© 11
AGC
DSP Vector space …
Professor A G Constantinides© 12
AGC
DSP Linear Independence
If there is no nonzero set of coefficients
{ci } i 1,2,3,4,...
m
such that ci pi [ n] 0
i 1
is linearly dependent
Professor A G Constantinides© 13
AGC
DSP Linear Independence
Examples:
2 3 4
T
1) p1
p 2 1 6 2
T
p 3 1 6 2
T
Observe that
4p1 5p 2 3p 3 0
Professor A G Constantinides© 14
AGC
DSP Linear Independence
Examples
2) p1 t
p2 1 t
Professor A G Constantinides© 15
AGC
DSP Linear Independence
Professor A G Constantinides© 16
AGC
DSP The Span of Vectors
The Span of a vector set is the set of all possible
vectors that can be reached (ie made through linear
combinations) from the given set.
Professor A G Constantinides© 17
AGC
DSP The Span of Vectors
Example: The vectors below are in 3-D real vector
space.
p1 1 1 0
T
p 2 0 1 0
T
x c1 c1 c2 0
T
Professor A G Constantinides© 18
AGC
DSP Basis and Dimension
If { pi [ n]} i 1,2,3,4,... is a selection of
linearly independent vectors from a vector space
such that they span the entire space then we say the
selected vectors form a (complete) basis.
The number of elements in the basis is the cardinality
of the basis
The least number of independent vectors to span a
given vector space S is called the dimension of the
vector space, usually designated as
dim(S )
Professor A G Constantinides© 19
AGC
DSP IMPORTANT!
Professor A G Constantinides© 20
AGC
DSP
Basis
Professor A G Constantinides© 21
AGC
DSP Vector Spaces
Let us start with the intuitive notion that we can
represent a signal as
x[n] ci pi [n]
i 0
This representation is called a projection of x[n],the
signal, into the linear vector space
{ pi [n]} i 1,2,3,4,...
The vectors above are linearly independent and can
span any signal in the space
Professor A G Constantinides© 22
AGC
DSP Vector Spaces
Examples are seen in Matrix Theory and typically in
Fourier Analysis.
Professor A G Constantinides© 23
AGC
DSP Norm
A real valued function x is the norm of x when
it satisfies
Positivity x 0
Scaling x x
Triangle inequality xy x y
Professor A G Constantinides© 24
AGC
Induced norm/Cauchy
DSP Schwartz inequality
*
x, y x, y
x, y y , x
Professor A G Constantinides© 26
AGC
DSP Inner Product
x y , z x, z y , z
x, x 0 x 0
x, x 0 x 0
x, y xT y xi yi
Professor A G Constantinides© 27
AGC
DSP Direction of Vectors
From the two and three dimensional cases we define
the angle between two vectors to be given from
x, y
cos
x2y2
If y ax the vectors are colinear
Professor A G Constantinides© 28
AGC
DSP Orthonormal
A set of vectors [p1 , p 2 , p 3 ,...p m ] is
orthonormal when
p i , p j ij
(Pythagoras)
If x y then the indiced norms satisfy
2 2 2
xy x y
Professor A G Constantinides© 29
AGC
DSP Weighted inner product
Very often we want a weighted inner product which
we define as
H
x, y W
y Wx
Professor A G Constantinides© 30
AGC
DSP Example
Let
1 2
x1 x2
1 1
Clearly 2
xT1 x 2 1 1 3 0
while
1
2 2 2 2 2
2 2 x 2 1 1 2 2 1 0
T
x1
Professor A G Constantinides© 31
AGC
DSP Example
Note that the inner product in the previous example
cannot serve as a norm as for any
a
x
a
we have T
x Wx 0
This violates the conditions for a norm
Professor A G Constantinides© 32
AGC
DSP Complete spaces
If every signal in a signal space is reachable (ie can
be spanned) by a set of vectors then the space is
complete and the reaching a complete set
This means that there will be no left over part of a
given signal expressed as an appropriate linear
combination of basis vectors
For example a Fourier series reaches all periodic
signals in the least square sense, and hence the set
of all complex exponentials is a complete set
Professor A G Constantinides© 33
AGC
DSP Hilbert spaces
Complete linear vector spaces with induced l2
norms are known as Hilbert Spaces
In signal processing we are mainly interested in finite
energy signals ie in Hilbert spaces
If the norm above is orther than the second then we
have Banach Spaces. (Such norms are useful in
approximating signals and system functions as we
shall see later)
Professor A G Constantinides© 34
AGC
DSP Orthogonal subspaces
Let S be a inner product signal (vector) space and V
and W be subspaces of S.
Then V and W are orthogonal if for every v V
and w W we have v, w 0
In above the set of all vectors orthogonal to a
subspace V is called the orthogonal complement
of the of the subspace denoted by V
Professor A G Constantinides© 35
AGC
DSP Inner-sum spaces
Let S be a inner product signal (vector) space and V
and W be subspaces of S with v V and w W
Then V +W is the inner sum space that consists of
all vectors x w v
Professor A G Constantinides© 37
AGC
DSP Example …
The orthogonal complement of A is found by
observing the the second member of the set is
orthogonal to all vectors having as their first entry a
zero. Thus
A 0 0 0, 0 1 0, 0 0 1, 0 1 1
Thus observe that B A
And the pair-wise sum
A B 0 0 0, 0 1 0, 1 0 0, 1 1 0
Professor A G Constantinides© 38
AGC
DSP Disjoint spaces
If two linear vector spaces of the same dimensionality
have only the zero vector in common they are called
disjoint.
Two disjoint spaces are such that one is the algebraic
complement of the other
Their sum is the entire vector space
Professor A G Constantinides© 39
AGC
DSP Disjoint spaces …
Let S be a inner product signal (vector) space and V
and W be subspaces of S
Professor A G Constantinides© 40
AGC
DSP Projections
From above a pictorial representation can be
produced as
V
v x
W
w
S
Professor A G Constantinides© 41
AGC
DSP Projections
Let S be a inner product signal (vector) space and V
and W be subspaces of S
We can think of v V and w W as being the
projections of x w v in the component sets.
P : S V
We introduce the projection operator
x S we have
Professor A G Constantinides© 42
AGC
DSP Projections
Thus if x is already in V the operation does not
change the value of x
Thus
P(Px) Px
This gives us the definition
A linear tranformation is a projection if
P
P2 P
(Idempotent operator)
Professor A G Constantinides© 43