Scribd
Upload
42 views

Week 10 - Introduction To Random Processes

1) A random process is a family of random variables indexed by time or space that evolves randomly. It maps the outcome of a random experiment to a waveform that is a function of time. 2) Properties of interest for random processes include the mean, autocorrelation, and autocovariance. A wide-sense stationary process has a constant mean and an autocorrelation that depends only on time difference. 3) The power spectral density is the frequency domain representation of a random process and is obtained by taking the Fourier transform of the autocorrelation function. It provides important information about the process.

Uploaded by

Haris Ghafoor
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
42 views

Week 10 - Introduction To Random Processes

1) A random process is a family of random variables indexed by time or space that evolves randomly. It maps the outcome of a random experiment to a waveform that is a function of time. 2) Properties of interest for random processes include the mean, autocorrelation, and autocovariance. A wide-sense stationary process has a constant mean and an autocorrelation that depends only on time difference. 3) The power spectral density is the frequency domain representation of a random process and is obtained by taking the Fourier transform of the autocorrelation function. It provides important information about the process.

Uploaded by

Haris Ghafoor
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

MATH-361

Probability and Statistics

Lecture 19

Introduction to Random Processes

Kamran Aziz Bhatti | Assistant Professor | Dept. of Electrical Engg. | NUST College of Electrical & Mechanical Engineering | Pakistan
Random Processes

• Numerical quantities that evolve randomly in time or space


• Indexed family of random variables
• Indexing is with respect to time or space
• 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 that is function of time

2
Random Processes

3
Mean, Autocorrelation & Autocovariance

4
Mean, Autocorrelation & Autocovariance

• Stationary to first order


• Distribution function (cdf) and density function (pdf) are not
function of time (strictly stationary)
• Mean of the random process is constant (wide-sense stationary)

• Stationary to second order


• Autocorrelation of random process depends only on the time
difference (wide-sense stationary)

• Autocovariance can be found using mean and autocorrelation

5
Mean, Autocorrelation & Autocovariance

• Mean and autocorrelation function only provide a partial


description of the distribution of random process
• All strictly stationary processes are wide sense stationary but
not all wide sense stationary processes are strictly stationary
• Considering wide sense stationary process, autocorrelation can
be written as
𝑅𝑋 𝑡2 − 𝑡1 = 𝑅𝑋 𝜏 = 𝐸[𝑋 𝑡 + 𝜏 𝑋(𝑡)]

6
Properties of Autocorrelation

• The mean square value of the process may be obtained from


autocorrelation by putting 𝜏 = 0
• The autocorrelation function is an even function of time
difference 𝜏
• The autocorrelation function has its maximum magnitude at
𝜏=0
• Describes the “interdependence” of two random variables

7
Cross-Correlation

• Considering both X and Y as wide sense stationary processes,


cross-correlation can be written as
• 𝑅𝑋𝑌 𝑡2 − 𝑡1 = 𝑅𝑋𝑌 𝜏 = 𝐸 𝑋 𝑡 + 𝜏 𝑌 𝑡
• Quadrature modulated processes
8
Random Processes & LTI Systems

• Effect of passing a random process through LTI system


• Mean, autocorrelation
• cdf, pdf
• Stationarity

• If input X(t) is WSS, Mean of the output is just equal to the


product of mean of the input and dc response of the system
• If the input to LTI system is WSS, the output of the LTI system is
also a WSS random process

9
Power Spectral Density (PSD)

• Frequency domain representation of random processes


• A single realization cannot be used to take FT
• ACF is used, being a statistical average
• The FT of the ACF is called the Power Spectral Density 𝑺𝑿 (𝒇)
of the random process X(t)
• The PSD and ACF of a WSS random process form a FT pair with
f and 𝜏 as the variables of interest.

10
Power Spectral Density (PSD)

• Basic relations in the theory of spectral analysis


∞
𝑆𝑋 𝑓 = 𝑅𝑋 𝜏 exp −𝑗2𝜋𝑓𝜏 𝑑𝜏
−∞
∞
𝑅𝑋 𝜏 = 𝑆𝑋 𝑓 exp 𝑗2𝜋𝑓𝜏 𝑑𝑓
−∞

• These relations show that if either the ACF or the PSD of a


random process is known, the other can be found

11
Properties of PSD

• Property 1: The zero frequency value of the PSD of a WSS


process is the total area under the graph of the ACF.
• Property 2: The mean square value of a WSS process equals
the total area under the graph of the PSD
• Property 3: The PSD of a WSS process is always nonnegative
• Amplitude spectrum of X(t)

• Property 4: The PSD of a real valued process is an even


function of frequency

12
Gaussian Process & its Properties

• Property 1: If a Gaussian process is applied to a stable linear filter,


the random process at the output of the filter is also Gaussian

• Property 2: Set of random variables obtained by sampling a


Gaussian random process are jointly Gaussian and this joint Gaussian
pdf is completely specified by mean vector and covariance matrix.

• Property 3: If a Gaussian process is WSS, then the process is also


strictly stationary

• Property 4: If set of random variables obtained by sampling a


Gaussian random process are uncorrelated, then they are also
independent.

13

You might also like