Linear Time-Invariant (LTI) Systems with Random Inputs

Linear Time-Invariant (LTI) Systems:

A linear time-invariant (LTI) system can be represented by its impulse response (Figure 10.6). More specifically, if
X(t) is the input signal to the system, the output, Y (t), can be written as
∞ ∞

Y (t) = ∫ h(α)X(t − α) dα = ∫ X(α)h(t − α) dα.

−∞ −∞

The above integral is called the convolution of h and X , and we write

Y (t) = h(t) ∗ X(t) = X(t) ∗ h(t).

Note that as the name suggests, the impulse response can be obtained if the input to the system is chosen to be the unit
impulse function (delta function) x(t) = δ(t). For discrete-time systems, the output can be written as (Figure 10.6)

Y (n) = h(n) ∗ X(n) = X(n) ∗ h(n)

∞ ∞

= ∑ h(k)X(n − k) = ∑ X(k)h(n − k).

k=−∞ k=−∞

The discrete-time unit impulse function is defined as

⎧ 1 n = 0
⎪

δ(n) = ⎨
⎩
⎪
0 otherwise

For the rest of this chapter, we mainly focus on continuous-time signals.

Figure 10.6 - LTI systems.

LTI Systems with Random Inputs:

Consider an LTI system with impulse response h(t). Let X(t) be a WSS random process. If X(t) is the input of the
system, then the output, Y (t), is also a random process. More specifically, we can write
 Y (t) = h(t) ∗ X(t)
∞

= ∫ h(α)X(t − α) dα.
−∞

Here, our goal is to show that X(t) and Y (t) are jointly WSS processes. Let's first start by calculating the mean
function of Y (t), μY (t). We have

μY (t) = E [Y (t)] = E [∫ h(α)X(t − α) dα]

−∞
∞

= ∫ h(α)E [X(t − α)] dα

−∞
∞

= ∫ h(α)μX dα
−∞
∞

= μX ∫ h(α) dα.
−∞

We note that μY (t) is not a function of t, so we can write

∞

μY (t) = μY = μX ∫ h(α) dα.

−∞

Let's next find the cross-correlation function, RXY (t1 , t2 ) . We have

∞

RXY (t1 , t2 ) = E [X(t1 )Y (t2 )] = E [X(t1 ) ∫ h(α)X(t2 − α) dα]

−∞

= E [∫ h(α)X(t1 )X(t2 − α) dα]

−∞
∞

= ∫ h(α)E [X(t1 )X(t2 − α)] dα

−∞
∞

= ∫ h(α)RX (t1 , t2 − α) dα
−∞
∞

= ∫ h(α)RX (t1 − t2 + α) dα (since X(t) is WSS).

−∞

We note that RXY (t1 , t2 ) is only a function of τ = t1 − t2 , so we may write

∞

RXY (τ ) = ∫ h(α)RX (τ + α) dα
−∞

= h(τ ) ∗ RX (−τ ) = h(−τ ) ∗ RX (τ ).

Similarly, you can show that

RY (τ ) = h(τ ) ∗ h(−τ ) ∗ RX (τ ).

This has been shown in the Solved Problems section. From the above results we conclude that X(t) and Y (t) are
jointly WSS. The following theorem summarizes the results.
Theorem
Let X(t) be a WSS random process and Y (t) be given by

Y (t) = h(t) ∗ X(t),

where h(t) is the impulse response of the system. Then X(t) and Y (t) are jointly WSS. Moreover,
∞
1. μY (t) = μY = μX ∫
−∞
h(α) dα;
∞
2. RXY (τ ) = h(−τ ) ∗ RX (τ ) = ∫
−∞
h(−α)RX (t − α) dα ;

3. RY (τ ) = h(τ ) ∗ h(−τ ) ∗ RX (τ ).

Frequency Domain Analysis:

Let's now rewrite the statement of Theorem 10.2 in the frequency domain. Let H (f ) be the Fourier transform of h(t),
∞
−2jπf t
H (f ) = F {h(t)} = ∫ h(t)e dt.
−∞

H (f ) is called the transfer function of the system. We can rewrite

∞

μY = μX ∫ h(α) dα
−∞

as

μY = μX H (0)

Since h(t) is assumed to be a real signal, we have

∗
F {h(−t)} = H (−f ) = H (f ),

where ∗ shows the complex conjugate. By taking the Fourier transform from both sides of
RXY (τ ) = RX (τ ) ∗ h(−τ ) , we conclude

∗
SXY (f ) = SX (f )H (−f ) = SX (f )H (f ).

Finally, by taking the Fourier transform from both sides of RY (τ ) = h(τ ) ∗ h(−τ ) ∗ RX (τ ), we conclude
∗
SY (f ) = SX (f )H (f )H (f )

2
= SX (f )|H (f )| .

2
SY (f ) = SX (f )|H (f )|

Example
Let X(t) be a zero-mean WSS process with RX (τ ) = e
−|τ |
. X(t) is input to an LTI system with
−−−−− −−−
⎧ √1 + 4π 2 f 2 |f | < 2
⎪

|H (f )| = ⎨
⎩
⎪
0 otherwise
Let Y (t) be the output.

a. Find μY (t) = E [Y (t)] .

b. Find RY (τ ).
c. Find E [Y (t)2 ] .

Solution
Note that since X(t) is WSS, X(t) and Y (t) are jointly WSS, and therefore Y (t) is WSS.
a. To find μY (t), we can write

μY = μX H (0)

= 0 ⋅ 1 = 0.

b. To find RY (τ ), we first find SY (f ) .

2
SY (f ) = SX (f )|H (f )| .

From Fourier transform tables, we can see that

−|τ |
SX (f ) = F {e }

2
= .
2
1 + (2πf )

Then, we can find SY (f ) as

2
SY (f ) = SX (f )|H (f )|

⎧ 2 |f | < 2
⎪

= ⎨
⎩
⎪
0 otherwise

We can now find RY (τ ) by taking the inverse Fourier transform of SY (f ) .

RY (τ ) = 8sinc(4τ ),

where

sin(πf )
sinc(f ) = .
πf

c. We have
2
E [Y (t) ] = RY (0) = 8.