You are on page 1of 12

Review of Signals & Systems: Part 2

and Channel Models

Satyajit Thakor
IIT Mandi
Systems

I Systems are operators which produce output for a given input.


I Discrete-time and continuous-time systems
I Linear and nonlinear systems: A system T is linear iff it satisfies

T [x1 (t) + x2 (t)] = T [x1 (t)] + T [x2 (t)] (additivity)


T [αxi (t)] = αT [xi (t)] (homogeneity)

otherwise it is nonlinear where, xi (t) are inputs and α is a scalar.


I i.e., response to linear combinations of inputs is equal to the
linear combination of the response to the inputs.
Systems

I Time-invariant and time-varying systems: Let y(t) be the


response of a system to x(t). This system is time-invariant iff
for all x(t) and all t0 , its response to x(t − t0 ) is y(t − t0 ).
I For linear time-invariant systems, the response of the system
can be calculated by finding convolution of the input x(t) and
impulse response h(t) = T [δ(t)] of the system.
I Prove (homework)
I For linear time-varying systems, impulse response is time
varying (not fixed).
I Causal and noncausal systems: A system is causal if
y(t0 ) = T [x(t) : t ≤ t0 ] for every time t0 .
I Prove (homework): An LTI system is causal iff h(t) is causal.
Additive noise channel

I We will discuss time-domain representation of three channel


models.
I Transmitted signal s(t) is distorted by additive noise n(t) (e.g.,
thermal noise), the received signal is
r(t) = s(t) + n(t).

I If the transmitted signal is further attenuated by a factor a(a


different kind of distortion) during propagation then the
received signal is
r(t) = as(t) + n(t).
Linear time-invariant filter channel

I Some channels like wireline telephone channels use filters to


restrict a signal to assigned bandwidth and thus do not interfere
with other signals at adjacent frequencies. Generally, the
wireline channel properties are “almost” time-invariant.
I Such channels are modeled as linear time-invariant filter
channels. Example: coaxial cable channel

Z ∞
r(t) = s(t) ∗ h(t) + n(t) = h(τ )s(t − τ )dτ + n(t)
Z−∞

= s(τ )h(t − τ )dτ + n(t)
−∞
Linear time-invariant filter channel

I An example of a multipath linear time-invariant filter wireless


channel channel:
I Consider the L-path LTI channel described by

L
X
h(t) = ak δ(t − τk )
k=1

I Pictorial representation and description of the channel


I If the input signal to this channel is s(t) then the output of the
channel (excluding the additive noise component) is
L
X
r(t) = ak s(t − τk )
k=1

I Proof:
Linear time-invariant filter channel

I Is the additive noise channel r(t) = as(t) + n(t) an LTI filter


channel?
Linear time-variant filter channel

I Underwater acoustic channels, ionospheric radio channels,


mobile wireless channels suffer multipath propagation.
I These are time varying due to Rx/Tx movement and/or change
in channel conditions. Example.

Z ∞
r(t) = s(t) ∗ h(τ, t) + n(t) = s(t − τ )h(τ, t)dτ + n(t)
−∞

I h(τ ; t) is the response of the channel at time t, due to an


impulse applied at time t − τ .
Linear time-variant filter channel

I Show that h(τ ; t) is the response of the channel (excluding the


noise component) at time t, due to an impulse applied at time
t − τ.
I Proof (students may skip it):
Revisiting complex exponentials (part 2)

I The set of complex exponentials are the eigenfunctions of LTI


systems.
I That is, applying the convolution operator to a complex
exponential function results in scalar multiple of the complex
exponential.
I Let x(t) = Aej(2πf0 t+θ) be the input to an LTI system. Then,
Z ∞
y(t) = h(τ )Aej2πf0 (t−τ )+jθ dτ
−∞
Z ∞
= Aej(2πf0 t+θ) h(τ )e−j2πf0 τ dτ
−∞
Z ∞
h(t) ∗ x(t) = h(τ )e−j2πf0 τ dτ x(t)
−∞

I meaning...
Fourier series

I Fourier series expansion: A periodic signal with period T0 = 1


f0
can be expanded in terms of complex exponential signals as

X
x(t) = xn ej2πnf0 t
−∞

Z α+T0
1
where xn = x(t)e−j2πnf0 t dt
T0 α
if Dirichlet conditions, x(t) in the period (1) is absolutely
integrable (2) has finite number of extrema and (3) has finite
discontinuities, are satisfied.
Homework

I Revise the basic concepts in real analysis, linear algebra, signals


and systems etc.
I For example: differentiation, integration, limit, sequences, linear
operators, trigonometric functions and identities, Cartesian,
polar coordinate systems
I Solve problems on signals and systems given in reference books.

I Reading:
Chapters 1 & 2 of J. G. Proakis and M. Salehi, Fundamentals of
Communication Systems

You might also like