You are on page 1of 9

Aims and Objectives

X By the end of the course, you would have understood:


• Basic signal analysis (mostly continuous-time)
E 2.5 • Basic system analysis (also mostly continuous systems)
• Time-domain analysis (including convolution)
Signals & Linear Systems
• Laplace Transform and transfer functions
• Fourier Series (revision) and Fourier Transform
• Sampling Theorem and signal reconstructions
Peter Cheung
Department of Electrical & Electronic Engineering • Basic z-transform
Imperial College London
(Based on course by Tania Stathaki)

URL: www.ee.ic.ac.uk/pcheung/
E-mail: p.cheung@ic.ac.uk

PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 1 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 2

About the course A demonstration …..

X Lectures - around 9 weeks (15-17 hours) X This is what you will be able to do in your 3rd year (helped by this course)
X Study Groups – 1 hr per week X You will be able to design and implement a NOISE CANCELLING system
X Assessment – 100% examination in June
X Handouts in the form of PowerPoint slides
X Text Book
• A. Oppenheim, A. Wilsky, S. Nawab, “Signals and Systems” (original
recommended text), Prentice Hall (~£49) (Famous and standard textbook,
used on many courses)
• B.P. Lathi, “Linear Systems and Signals”, 2nd Ed., Oxford University
Press (~£36) (MY PREFERRED TEXTBOOK – I WILL BE FOLLOWING
THIS CLOSELY. IN MY VIEW, THIS IS A CHEAPER BOOK, HARDBACK,
AND EXPLAINS THE UNDERLINING CONCEPT BETTER.)
X Notes from original course by Dr Stathaki on
• www.commsp.ee.ic.ac.uk/~tania/

PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 3 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 4
Examples of signals (1)

X Electroencephalogram (EEG) signal (or brainwave)

Lecture 1

Basics about Signals


(Lathi 1.1-1.5; Oppenheim 1.0-1.4)

Peter Cheung
Department of Electrical & Electronic Engineering
Imperial College London

URL: www.ee.imperial.ac.uk/pcheung/
E-mail: p.cheung@imperial.ac.uk

PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 5 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 6

Examples of signals (2) Examples of signals (3)

X Stock Market data as signal (time series) X Magnetic Resonance Image (MRI) data as 2-dimensional signal

PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 7 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 8
Size of a Signal x(t) (1) Size of a Signal x(t) (2)

X Measured by signal energy Ex: X If amplitude of x(t) does not → 0 when t → ∞, need to measure power Px instead:

X Generalize for a complex valued signal to: X Again, generalize for a complex valued signal to:

X Energy must be finite, which means Lathi


Section 1.1

L1.1 L1.1

PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 9 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 10

Size of a Signal x(t) (3) Useful Signal Operations –Time Shifting (1)

X Signal with finite energy (zero power) X Signal may be delayed by time T:

X or advanced by time T:

X Signal with finite power (infinite energy)


φ (t – T) = x (t)

L1.1 L1.2.1
PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 11 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 12
Useful Signal Operations –Time Scaling (2) Useful Signal Operations –Time Reversal (3)

X Signal may be compressed in time (by a X Signal may be reflected about the
factor of 2): vertical axis (i.e. time reversed):

X or expanded in time (by a factor of 2): X We can combine these three


operations.
X For example, the signal x(2t - 6) can
be obtained in two ways;
φ (2t ) = x(t ) • Delay x(t) by 6 to obtain x(t - 6),
and then time-compress this
signal by factor 2 (replace t with
X Same as recording played back at 2t) to obtain x (2t - 6).
twice and half the speed • Alternately, time-compress x (t)
respectively by factor 2 to obtain x (2t), then
delay this signal by 3 (replace t
with t - 3) to obtain x (2t - 6).
L1.2.2 L1.2.3
PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 13 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 14

Signals Classification (1) Signal Classification (2) – Continuous vs Discrete

X Signals may be classified into:


1. Continuous-time and discrete-time signals X Continuous-time
2. Analogue and digital signals
3. Periodic and aperiodic signals
4. Energy and power signals
5. Deterministic and probabilistic signals
6. Causal and non-causal
7. Even and Odd signals
X Discrete-time

L1.3 L1.3
PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 15 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 16
Signal Classification (3) – Analogue vs Digital Signal Classification (4) – Periodic vs Aperiodic

X A signal x(t) is said to be periodic if for some positive constant To


Digital, continuous

Analogue, continuous
X The smallest value of To that satisfies the periodicity condition of this
equation is the fundamental period of x(t).

Analogue, discrete Digital, discrete

L1.3 L1.3
PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 17 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 18

Signal Classification (5) – Deterministic vs Random Signal Classification (6) – Causal vs Non-causal

Deterministic
Causal

Non-causal
Random

L1.3
PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 19 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 20
Signal Classification (7) – Even vs Odd Signal Models (1) – Unit Step Function u(t)

X Step function defined by:


Even

X Useful to describe a signal that begins


at t = 0 (i.e. causal signal).
Odd X For example, the signal
represents an everlasting exponential
that starts at t = -∞.
X The causal for of this exponential can
be described as:

L1.4.1
PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 21 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 22

Signal Models (2) – Pulse signal Signal Models (3) – Unit Impulse Function δ(t)

X A pulse signal can be presented by two step functions: X First defined by Dirac as:

Approximation of
Unit Impulse an Impulse

L1.4.1 L1.4.2
PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 23 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 24
Signal Models (4) – Unit Impulse Function δ(t) Multiplying a function Φ(t) by an Impulse

X May use functions other than a rectangular pulse. Here are three X Since impulse is non-zero only at t = 0, and Φ(t) at t = 0 is Φ(0), we get:
example functions:
X Note that the area under the pulse function must be unity

Exponential Triangular Gaussian X We can generalise this for t = T:

L1.4.2 L1.4.2
PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 25 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 26

Sampling Property of Unit Impulse Function The Exponential Function est (1)

X Since we have: X This exponential function is very important in signals & systems, and the
parameter s is a complex variable given by:
X It follows that:

X This is the same as “sampling” Φ(t) at t = 0.


X If we want to sample Φ(t) at t = T, we just multiple Φ(t) with

X This is called the “sampling or sifting property” of the impulse.

L1.4.2 L1.4.3
PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 27 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 28
The Exponential Function est (2) The Exponential Function est (2)

X If σ = 0, then we have the function , which has a real


frequency of ω
X Therefore the complex variable s = σ + jω is the complex
frequency
X The function est can be used to describe a very large class of
signals and functions. Here are a number of example:

L1.4.3 L1.4.3
PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 29 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 30

The Complex Frequency Plane s = σ + jω Even and Odd functions (1)

+jω The s-plane X A real function xe(t) is said to be an even function of t if

s on y-axis
s on right of y-axis

-σ +σ
X A real function xo(t) is said to be an odd function of t if

s on left of y-axis

s on x-axis
-jω L1.4.3 L1.5
PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 31 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 32
Even and Odd functions (2) Even and Odd functions (3)

X Even and odd functions have the following properties: X Consider the causal exponential function
• Even x Odd = Odd
• Odd x Odd = Even
• Even x Even = Even

X Every signal x(t) can be expressed as a sum of even and


odd components because:

L1.5 L1.5
PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 33 PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 34

Relating this lecture to other courses

X The first part of this lecture on signals has been covered in


this lecture was covered in the 1st year Communications
course (lectures 1-3)
X This is mostly an introductory and revision lecture

PYKC 9-Mar-08 E2.5 Signals & Linear Systems Lecture 1 Slide 35