Ch-4 Random Process = Stochastic Process

Random process is a family of random variables or random signals. It is a collection (or ensemble) of
random functions representing the evolution of some system of random values over time. Another name for
random process is stochastic process. Stochastic is a Greek term which means random or guessing. As
random process is a probabilistic counterpart of deterministic process, we first understand what deterministic
process is?

Deterministic Process
It is a process where given the starting point, we can know or predict the behaviour of the process. Consider
the following process:

𝑥(𝑡) = [𝑥(𝑡 − 1)]2

𝑎𝑛𝑑 𝑥(0) = 𝑎 (𝑠𝑜𝑚𝑒 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡) … . 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑜𝑟 𝑠𝑡𝑎𝑟𝑡𝑖𝑛𝑔 𝑝𝑜𝑖𝑛𝑡

𝑥(0) = 𝑎 (𝑔𝑖𝑣𝑒𝑛)

𝑥(1) = [𝑥(1 − 1)]2 = 𝑥(0)2 = 𝑎2

𝑥(2) = [𝑥(2 − 1)]2 = 𝑥(1)2 = (𝑎2 )2 = 𝑎4

𝑥(3) = [𝑥(3 − 1)]2 = 𝑥(2)2 = (𝑎4 )2 = 𝑎8

So, the process goes as follows: 𝑎, 𝑎2 , 𝑎4 , 𝑎8 , … … … ….

If we know starting point, we will know the rest of the sequence. Without any ambiguity (doubt). This is the
deterministic process. In a deterministic process, each subsequent step is said to be known with probability 1
(complete certainty). Another example of deterministic process is a sinusoidal process. No uncertainty about
its time-dependent behaviour at any instant of time. But many real-world problems are random in nature &
hence random processes.

Random Process
The value of a random signal can’t be predicted ahead of time. There are many examples of random signals.:
speech & music signals, output voltage of a SOLAR DETECTOR which depends on weather conditions, the
marks you score in university examination etc.

A stochastic process is simply a collection of

random variables i.e., random signals indexed by
time (t). if the index (t) is continuous, it is regarded
as continuous-time process. If ‘t’ is discrete, we call
discrete-time process.

Continuous-time

Time
Discrete-time

1|Page
The notion of stochastic process is very important in mathematics theory and its applications in science &
technology. It is used to model a large number of various phenomena where the quantity of interest varies
discretely or continuously through time in non-predictable manner. Models are mathematical representation
of real systems & fundamentals tools in science and Engineering.

Random processes have 2 fundamental properties:

 They are functions of time

 They are random i.e., before conducting the experiment, it is not possible to define the waveforms
that will be observed in the future.

Coin tossing is a random process with few number of random variables like:
X = getting head
Y = getting tail
Z = number of heads – number of tails etc

Some real-life examples of random processes are:

 Prices of a stock market

 Sum of series of die rolls
 Velocity of wind
 Customers entering a store
 Incoming jobs for a computer system
 Temperature variation/climate variation etc

 For a random variable, the outcome of a random experiment is mapped into a number
 For a random process, the outcome of a random experiment is mapped into a waveform

Example: 5 dice are thrown

Let X = RV = ‘sum of all the faces’

Xmin = 1+1+1+ 1+1 = 5

Xmax = 6+6+6+ 6+6 = 30

Each sample point in the sample space can be mapped to one random waveform (or signal). In above figure
sample point 30 is mapped to X (t, s1) and sample point 7 mapped to X (t, sk) and so on. Above sample space
contains 26 sample points will be present. So, this Random Process consists of 26 random waveforms.

2|Page
𝑋(𝑡, 𝑠1 ) = 𝑃𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑜𝑓 𝑜𝑐𝑐𝑢𝑟𝑒𝑛𝑐𝑒 𝑜𝑓 𝑠1

𝑋(𝑡, 𝑠2 ) = 𝑃𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑜𝑓 𝑜𝑐𝑐𝑢𝑟𝑒𝑛𝑐𝑒 𝑜𝑓 𝑠2

𝑋(𝑡, 𝑠3 ) = 𝑃𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑜𝑓 𝑜𝑐𝑐𝑢𝑟𝑒𝑛𝑐𝑒 𝑜𝑓 𝑠3

Correlation
Correlation is a measure of similarity between two signals. Two signals mean two random waveforms.
There are two types of correlation

Cross Correlation

Correlation
Auto Correlation

 If two random signals come from same random process, auto-correlation function is defined.

 If two random signals come from different random processes, it will be called the cross-correlation
function.

Note that correlation between 2 random variables X and Y is the expected value of their product i.e., E[XY]
= correlation. Correlation depends on how rapidly the value of the random variable can change with respect
to time (t).

Correlation is the measure of similarity between two waveforms.

NOTE:

Statistical measures can be used to characterise random signals or processes. Statistical measures in time-
domain include mean, variance, auto-correlation, correlation etc. power spectral density (PSD) is used to
characterise random process in frequency domain.

Q. Find the covariance and correlation between random variables X and Y, where X = number of study
hours, Y = number of sleeping hours. Below data is given:

X 2 4 6 8 10
Y 10 9 8 7 6

Sol:

Covariance between X&Y:

𝐶𝑜𝑣(𝑋, 𝑌) = ∑(𝑋 − 𝑚𝑥 )(𝑌 − 𝑚𝑦 )

3|Page
 2 + 4 + 6 + 8 + 10
𝑚𝑥 = 𝑚𝑒𝑎𝑛 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑅𝑉 𝑋 = =6
5

10 + 9 + 8 + 7 + 6
𝑚𝑦 = 𝑚𝑒𝑎𝑛 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑅𝑉 𝑌 = =8
5

X Y (𝑋 (𝑌 − 𝑚𝑦 ) (𝑋 − 𝑚𝑥 )(𝑌 − 𝑚𝑦 ) (𝑋 − 𝑚𝑥 )2 (𝑌 − 𝑚𝑦 )2
− 𝑚𝑥 )
2 10 2-6 = -4 10-8 = 2 -8 16 4
4 9 4-6 = -2 9-8 = 1 -2 4 1
6 8 6-6 = 0 8-8 = 0 0 0 0
8 7 8-6 = 2 7-8 = -1 -2 4 1
10 6 10-6 = 4 6-8 = -2 -8 16 4
∑ 𝑿 = 𝟑𝟎 ∑ 𝒀 = 𝟒𝟎 ∑(𝑿 − 𝒎𝒙 )(𝒀 − 𝒎𝒚 ) = −𝟐𝟎

𝑉𝑎𝑟(𝑋) = ∑(𝑋 − 𝑚𝑥 )2 = 16 + 4 + 0 + 4 + 16 = 40

2
𝑉𝑎𝑟(𝑌) = ∑(𝑌 − 𝑚𝑦 ) = 4 + 1 + 0 + 1 + 4 = 10

∴ 𝐶𝑜𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒 𝑏𝑒𝑡𝑤𝑒𝑒𝑛 𝑋 & 𝑌 = 𝐶𝑜𝑣(𝑋, 𝑌) = ∑(𝑋 − 𝑚𝑥 )(𝑌 − 𝑚𝑦 ) = −20

𝐶𝑜𝑣(𝑋,𝑌) −20 −20

And 𝐶𝑜𝑟𝑟𝑒𝑙𝑎𝑡𝑖𝑜𝑛 𝑏𝑒𝑡𝑤𝑒𝑒𝑛 𝑋 & 𝑌 = 𝐶𝑜𝑟𝑟(𝑋, 𝑌) = = = −1
√𝑉𝑎𝑟 (𝑋) 𝑉𝑎𝑟(𝑌) √40 𝑥 10 20

NOTE: value of correlation lies between +1 and -1

+ve Correlation

Discrete RVs -ve Correlation

No Correlation

o Positive correlation: when the value of one variable (say X) increases with other (Y)
o Negative correlation: X increase Y decrease
o No correlation: there is no relation between X and Y

4|Page
Q. Find the mean and variance, if they exist in each of the following distributions:
𝑥3
, 0<𝑥<1
(i) 𝑓(𝑥) = {12
0 , 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

𝑥
, 𝑥 = 1,2,3,4,5
(ii) 𝑓(𝑥) = {15
0 , 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

Sol:

(i)

𝑥3
𝑓(𝑥) = 𝑖𝑠 𝑎 𝑐𝑜𝑛𝑡𝑖𝑛𝑢𝑜𝑢𝑠 𝑑𝑖𝑠𝑡𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛
3
∞ 1 1
𝑥3 1 1
𝑚𝑒𝑎𝑛 = 𝐸[𝑋] = ∫ 𝑥 . 𝑓(𝑥)𝑑𝑥 = ∫ 𝑥 . = ∫ 𝑥 4 𝑑𝑥 =
12 12 60
−∞ 0 0

𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒, 𝜎 2 = 𝐸[𝑋 2 ] − 𝑚2
∞ 1
2] 2
𝑥3 1 2
𝐸[𝑋 = ∫ 𝑥 𝑓(𝑥)𝑑𝑥 = ∫ 𝑥 𝑑𝑥 =
12 72
−∞ 0

2 2]
12
1 2
∴ 𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒, 𝜎 = 𝐸[𝑋 − 𝑚 = − ( ) = 0.6136
72 60

(ii)

𝑥
𝑓(𝑥) = 𝑥 = 1,2,3,4,5 𝑖𝑠 𝑑𝑖𝑠𝑐𝑟𝑒𝑡𝑒 𝑑𝑖𝑠𝑡𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛
15
5 5 5
𝑥 1 1 11
𝑚𝑒𝑎𝑛 = 𝐸[𝑋] = ∑ 𝑥 . 𝑓(𝑥) = ∑ 𝑥 . = {∑ 𝑥 2 } = {1 + 4 + 9 + 16 + 15} =
15 15 15 13
1 1 1

𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒, 𝜎 2 = 𝐸[𝑋 2 ] − 𝑚2

5 5 5
2] 2
𝑥 1 2
𝐸[𝑋 = ∑ 𝑥 𝑓(𝑥) = ∑ 𝑥 = ∑ 𝑥 3 = 15
15 15
1 1 1

2 2] 2
11 2 14
∴ 𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒, 𝜎 = 𝐸[𝑋 − 𝑚 = 15 − ( ) =
3 9

5|Page
 𝑥 + 𝑦, 0 < 𝑥 ≤ 1
Q. Find the correlation coefficient of X & Y if 𝑓(𝑥) = {
0, 𝑒𝑙𝑠𝑒𝑤ℎ𝑒𝑟𝑒

Sol: This is example of continuous process

𝐶𝑜𝑣(𝑋, 𝑌)
𝑐𝑜𝑟𝑟𝑒𝑙𝑎𝑡𝑖𝑜𝑛 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝜌 =
√𝑉𝑎𝑟(𝑋) 𝑉𝑎𝑟(𝑌)

𝐶𝑜𝑣(𝑋, 𝑌) = 𝐸[𝑋𝑌] − 𝐸[𝑋}𝐸[𝑌]

𝐸[𝑋] = ∫ 𝑥 . 𝑓(𝑥)𝑑𝑥
−∞

∞ ∞

= ∫ ∫ 𝑥 . 𝑓(𝑥, 𝑦)𝑑𝑥 𝑑𝑦
−∞ −∞

1 1

= ∫ ∫ 𝑥 . (𝑥 + 𝑦)𝑑𝑥 𝑑𝑦 =
𝑥=0 𝑦=0

1 1 1 1
7
∫ ∫ 𝑥 2 𝑑𝑥 𝑑𝑦 + ∫ ∫ 𝑥𝑦 𝑑𝑥 𝑑𝑦 =
12
𝑥=0 𝑦=0 𝑥=0 𝑦=0

7
𝑠𝑖𝑚𝑖𝑙𝑎𝑟𝑙𝑦 𝐸[𝑌] = (𝑐𝑎𝑙𝑐𝑢𝑙𝑎𝑡𝑖𝑜𝑛𝑠 𝑙𝑒𝑓𝑡 𝑓𝑜𝑟 𝑠𝑡𝑢𝑑𝑒𝑛𝑡)
12

∞ ∞

𝐸[𝑋𝑌] = ∫ ∫ 𝑥𝑦 . 𝑓(𝑥, 𝑦)𝑑𝑥 𝑑𝑦

−∞ −∞

1 1
1
= ∫ ∫ 𝑥𝑦 (𝑥 + 𝑦)𝑑𝑥 𝑑𝑦 =
3
𝑥=0 𝑦=0

1 7 7 −1
𝐶𝑜𝑣(𝑋, 𝑌) = 𝐸[𝑋𝑌] − 𝐸[𝑋]𝐸[𝑌] = − . =
3 12 12 144

∞ 1 1
5
𝐸[𝑋 2 ] = ∫ 𝑥 2 𝑓(𝑥)𝑑𝑥 = ∫ ∫ 𝑥 2 (𝑥 + 𝑦)𝑑𝑥 𝑑𝑦 =
12
−∞ 𝑥=0 𝑦=0

6|Page
 5
𝑠𝑖𝑚𝑖𝑙𝑎𝑟𝑙𝑦, 𝐸[𝑌 2 ] = (𝑐𝑎𝑙𝑐𝑢𝑙𝑎𝑡𝑖𝑜𝑛𝑠 𝑙𝑒𝑓𝑡 𝑓𝑜𝑟 𝑠𝑡𝑢𝑑𝑒𝑛𝑡)
12

2]
5 7 2 11
𝑉𝑎𝑟(𝑋) = 𝐸[𝑋 − (𝐸[𝑋])2 = − ( ) =
12 12 144

2]
5 7 2 11
𝑉𝑎𝑟(𝑌) = 𝐸[𝑌 − (𝐸[𝑌])2 = − ( ) =
12 12 144

−1
𝐶𝑜𝑣(𝑋, 𝑌)
144 −1
𝑐𝑜𝑟𝑟𝑒𝑙𝑎𝑡𝑖𝑜𝑛 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝜌 = = =
√𝑉𝑎𝑟(𝑋) 𝑉𝑎𝑟(𝑌) 11 11
144

Q. Given two zero mean unity variance random variables X and Y with a correlation coefficient 𝜌 = 0.5
from a new random variable Z =(CX-3Y)2 – 2Y where C is a real valued constant. Find the value of C such
that E[Z] = 9

Sol:

Z =(CX-3Y)2 – 2Y

𝐸[𝑍] = 𝐸[(CX − 3Y)2 – 2Y ]

= 𝐸[𝐶 2 𝑋 2 + 9𝑌 2 − 6𝐶𝑋𝑌 − 2𝑌]

= 𝐶 2 𝐸[𝑋 2 ] + 9𝐸[𝑌 2 ] − 6𝐶𝐸[𝑋𝑌] − 2𝐸[𝑌]

Given that X and Y are zero mean & unity variance random variables

∴ 𝐸[𝑋] = 𝐸[𝑌] = 0 & 𝜎𝑥 2 = 𝜎𝑦 2 = 1

∴ 𝑠𝑢𝑏𝑠𝑡𝑖𝑡𝑢𝑡𝑖𝑛𝑔 𝑎𝑏𝑜𝑣𝑒 𝑣𝑎𝑙𝑢𝑒𝑠 𝑖𝑛 𝐸[𝑍], 𝑤𝑒 𝑔𝑒𝑡

𝐸[𝑍] = 𝐶 2 𝐸[𝑋 2 ] + 9𝐸[𝑌 2 ] − 6𝐶𝐸[𝑋𝑌] − 2𝐸[𝑌]

= 𝐶 2 𝐸[𝑋 2 ] + 9𝐸[𝑌 2 ] − 6𝐶𝐸[𝑋𝑌] − 0

= 𝐶 2 𝐸[𝑋 2 ] + 9𝐸[𝑌 2 ] − 6𝐶𝐸[𝑋𝑌] ……………… equation 1

Also given that correlation coefficient 𝜌 = 0.5

𝐶𝑜𝑣(𝑋, 𝑌) 𝐸[𝑋𝑌] − 𝐸[𝑋]𝐸[𝑌]

𝑤𝑒 𝑘𝑛𝑜𝑤 𝑡ℎ𝑎𝑡 𝜌 = =
√𝑉𝑎𝑟(𝑋) 𝑉𝑎𝑟(𝑌) √1 𝑥 1

∴ 0.5 = 𝐸[𝑋𝑌]

7|Page
𝜎𝑥 2 𝑓𝑜𝑟𝑚𝑢𝑙𝑎 = 𝐸[𝑋 2 ] − (𝐸[𝑋])2 = 𝐸[𝑋 2 ] = 1 (𝑏𝑒𝑐𝑎𝑖𝑢𝑠𝑒 𝜎𝑥 2 = 1 𝑔𝑖𝑣𝑒𝑛)

𝜎𝑦 2 𝑓𝑜𝑟𝑚𝑢𝑙𝑎 = 𝐸[𝑌 2 ] − (𝐸[𝑌])2 = 𝐸[𝑌 2 ] = 1 (𝑏𝑒𝑐𝑎𝑖𝑢𝑠𝑒 𝜎𝑦 2 = 1 𝑔𝑖𝑣𝑒𝑛)

Substituting all values into equation 1, we get

𝐸[𝑍] = 𝐶 2 𝐸[𝑋 2 ] + 9𝐸[𝑌 2 ] − 6𝐶𝐸[𝑋𝑌]

9 = 𝐶 2 (1) + 9(1) − 6𝐶(0.5)

𝐶 2 − 6𝐶𝑥0.5 = 0

𝐶 2 − 3𝐶 = 0

𝐶(𝐶 − 3) = 0

∴ 𝐶 = 0 𝑜𝑟 𝐶 = 3

o NOTE: two random processes X(t) and Y(t) are said to be uncorrelated if their cross-correlation is
equal to the product of their mean functions i.e., E[XY] = E[X]. E[Y]

o Two random processes X(t) and Y(t) are said to be orthogonal if their cross-correlation is zero.

8|Page
Autocorrelation
Correlation of a function with itself is known as autocorrelation. Autocorrelation is used to compare a signal
with a time-delayed 𝝉 version of itself. If the signal is periodic, then the signal will be perfectly correlated
with a version of itself [if the time delay is an integer number of periods].

Autocorrelation function can be used to detect non-randomness in data. It describes the general dependency
of x(t) with its value at a short time latter 𝑖. 𝑒. , 𝑥(𝑡 + 𝝉), where 𝝉 = 𝑡𝑖𝑚𝑒 𝑑𝑒𝑙𝑎𝑦.

Properties of ACF (Auto Correlation Function)

ACH and their properties play a useful role in the analysis of systems with random inputs.

1. ACF of a random process X(t) is defined as 𝑅𝑋𝑋 (𝜏) = 𝐸[𝑋(𝑡)𝑋(𝑡 + 𝜏)]

2. ACF is an even function of 𝝉 i. e. , 𝑅𝑋𝑋 (𝜏) = 𝑅𝑋𝑋 (−𝜏) i.e, mirror image

3. The mean-square value of the random process can always be obtained by setting 𝝉 = 𝟎
𝑖. 𝑒. , 𝑅𝑋𝑋 (0) = 𝐸[𝑋 2 ]

4. The largest value of auto-correlation function always occurs at 𝝉 = 𝟎

𝑖. 𝑒. , 𝑅𝑋𝑋 (0) ≥ 𝑅𝑋𝑋 (𝜏)

NOTE:

Auto-correlation is a characteristic of data in which the correlation between the values of the same signal or
waveform is measured.

Power Spectral Density (PSD)

PSD is used to characterise random processes in frequency domain. PSD is also known as power spectrum
or simply spectrum. Spectral analysis refers to estimating power content of the frequency component of a
signal.

𝐿𝑒𝑡 X(t) = random process

𝑅𝑋 (𝜏) = 𝑎𝑢𝑡𝑜𝑐𝑜𝑟𝑟𝑒𝑙𝑎𝑡𝑖𝑜𝑛 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛

𝑆𝑋 (𝑓) = 𝑝𝑜𝑤𝑒𝑟 𝑠𝑝𝑒𝑐𝑡𝑟𝑎𝑙 𝑑𝑒𝑛𝑠𝑖𝑡𝑦 (𝑃𝑆𝐷)

9|Page
Autocorrelation function and power spectral density are Fourier transform pair

F.T. pair
𝑅𝑋 (𝜏) 𝑆𝑋 (𝑓)

F.T pair means Fourier Transform pair. If we find Fourier Transform of 𝑅𝑋 (𝜏), we get 𝑆𝑋 (𝑓). And if we find
inverse Fourier transform of 𝑆𝑋 (𝑓), 𝑤𝑒 𝑤𝑖𝑙𝑙 𝑔𝑒𝑡 𝑅𝑋 (𝜏).

Fourier & Inverse Fourier transforms

Let x(t) is any signal
∞

𝐹. 𝑇. {𝑥(𝑡)} 𝑋(𝑤) = ∫ 𝑥(𝑡)𝑒 −𝑖𝜔𝑡 𝑑𝑡

−∞
OR
∞

𝐹. 𝑇. {𝑥(𝑡)} 𝑋(𝑓) = ∫ 𝑥(𝑡)𝑒 −𝑖2𝜋𝑓𝑡 𝑑𝑡

−∞

FT is used to convert time-domain signal into frequency domain

Inverse FT: IFT converts frequency domain signal into time-domain.

∞
1
𝑥(𝑡) = ∫ 𝑥(𝑤)𝑒 𝑖𝜔𝑡 𝑑𝑤
2𝜋
−∞

= ∫ 𝑥(𝑓)𝑒 𝑖2𝜋𝑓𝑡 𝑑𝑓
−∞

Mean-square value of random process X(t) = E[X2]. Note that mean-square value is the average power of a
random process.

Mean-square value of a random process X(t):

∞
1
𝐸[𝑋 2 ] = ∫ 𝑆𝑥 (𝑤)𝑑𝑤
2𝜋
−∞

= ∫ 𝑆𝑥 (𝑓)𝑑𝑓
−∞

𝑆𝑥 (𝑓) can be obtained from 𝑆𝑥 (𝑤) by substituting 𝑤 = 2𝜋𝑓, where w = angular frequency in rad/sec. For
example, a frequently occurring spectral density has the form:

10 | P a g e
 2𝑎
𝑆𝑥 (𝑤) = . The corresponding spectral density in terms of f(frequency) would be
𝑤 2 +𝑎2

2𝑎
𝑆𝑥 (𝑓) = .
(2𝜋𝑓)2 +𝑎2

Spectral density at origin using 𝑆𝑥 (𝑤) i.e., at w = 0

𝟐
𝑺𝒙 (𝒘 = 𝟎) = 𝑽𝒐𝒍𝒕𝟐 /𝑯𝒛
𝒂
at w = 1 rad/sec
2𝑎
𝑆𝑥 (𝑤 = 1) = 2
𝑉𝑜𝑙𝑡 2 /𝐻𝑧
1+𝑎
Spectral density at origin using 𝑆𝑥 (𝑓)

2
𝑆𝑥 (𝑓 = 0) = 𝑉𝑜𝑙𝑡 2 /𝐻𝑧
𝑎
2𝑎
𝑆𝑥 (𝑓 = 1) = 𝑉𝑜𝑙𝑡 2 /𝐻𝑧
1 + 𝑎2

Observe that both results are same.

NOTE:

The PSD of a process is a measure of how the average power of the process is distributed with respect to
frequency.

Q. Find the average power of the random process with the following power spectrum:
𝒘
𝟏− , |𝒘| < 𝟒𝝅
𝑺𝒙 (𝒘) = { 𝟒𝝅
𝟎, |𝒘| ≥ 𝟒𝝅

Sol:
∞
1
𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑝𝑜𝑤𝑒𝑟, 𝐸[𝑋 2 ] = ∫ 𝑆𝑥 (𝑤)𝑑𝑤
2𝜋
−∞
4𝜋
1 𝑤
= ∫ (1 − ) 𝑑𝑤
2𝜋 4𝜋
−4𝜋
4𝜋 4𝜋
1 𝑤
= { ∫ 𝑑𝑤 − ∫ 𝑑𝑤 }
2𝜋 4𝜋
−4𝜋 −4𝜋
4𝜋
1 1 𝑤2
= { 𝑤]4𝜋
−4𝜋 − ] }=4
2𝜋 4𝜋 2 −4𝜋

11 | P a g e
Q. What is the average power of the square wave shown below:
Sol:

In above periodic continuous-time signal average power is defined for one cycle. The period of the signal is
T.

𝑇
+1, 0<𝑡<
𝑥(𝑡) = { 2
𝑇
−1, <𝑡<𝑇
2
1 𝑇/2
Average power 𝑃 = 𝑇
∫−𝑇/2 𝑥 2 (𝑡)𝑑𝑡
OR
𝑇
1
𝑃 = ∫ 𝑥 2 (𝑡)𝑑𝑡
𝑇
0

𝑇/2 𝑇
1 1 1
∴𝑃= ∫ (1)2 𝑑𝑡 + ∫(−1)2 𝑑𝑡 = 𝑥𝑇=1
𝑇 𝑇 𝑇
0 𝑇/2

Q. What is the average power of the discrete – time signal shown in below figure.
\

Sol:

Average power of the discrete – time signal x[n] of fundamental period N is given by:

𝑵−𝟏
𝟏
𝑷= ∑ 𝒙𝟐 (𝒏)
𝑵
𝒏=𝟎

From figure N=8

12 | P a g e
 7
1 1
∴ 𝑃 = ∑ 𝑥 2 (𝑛) = {𝑥 2 [0] + 𝑥 2 [1] + 𝑥 2 [2] + ⋯ … … 𝑥 2 [7]}
8 8
𝑛=0

1 8
= {1 + 1 + 1 + 1 + 1 + 1 + 1 + 1} = = 1
8 8

Q. A stationary random process has power spectral density given by:

𝟏𝟎, 𝒂<𝒇<𝒃
𝑺𝒙 (𝒇) = {
𝟎, 𝒆𝒍𝒔𝒆𝒘𝒉𝒆𝒓𝒆

Find the mean-square value of the process (i) if a = 4 & b = 5 (ii) if a = 0 & b = 5

Q. A stationary random process has a two-sided spectral density given by

𝟐𝟒 𝑽𝒐𝒍𝒕𝟐
𝑺 𝒙 ( 𝒘) = .
𝒘𝟐 + 𝟏𝟔 𝑯𝒛

Find the mean-square value of the process.

13 | P a g e