Scribd
Upload
38 views15 pages

Ee891 Stochastic Systems

This document discusses the transformation of random variables (RVs). It explains that if a new RV Y is defined as a function g(X) of an existing RV X, the probability distribution of Y can be determined from the distribution of X and the function g. Specifically, the cumulative distribution function (CDF) FY(y) of Y is equal to the CDF FX(x) of X, where x are the roots of the equation y = g(x). Taking the derivative with respect to y yields the probability density function (PDF) fy(y) of Y in terms of the PDF fx(x) of X. Examples are provided to illustrate monotonically increasing and decreasing transformations. The general transformation formula is derived

Uploaded by

RafeyTahir
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
38 views15 pages

Ee891 Stochastic Systems

This document discusses the transformation of random variables (RVs). It explains that if a new RV Y is defined as a function g(X) of an existing RV X, the probability distribution of Y can be determined from the distribution of X and the function g. Specifically, the cumulative distribution function (CDF) FY(y) of Y is equal to the CDF FX(x) of X, where x are the roots of the equation y = g(x). Taking the derivative with respect to y yields the probability density function (PDF) fy(y) of Y in terms of the PDF fx(x) of X. Examples are provided to illustrate monotonically increasing and decreasing transformations. The general transformation formula is derived

Uploaded by

RafeyTahir
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 15

EE891 STOCHASTIC

SYSTEMS

Lecture 10
Transformation of RVs
Transformation of RVs
◦ Consider a RV 𝑋 with a PDF 𝑓 𝑥 and CDF 𝐹 𝑥
◦ Define a new RV 𝑌: 𝑌 𝑔 𝑋 for some function 𝑔 , eg

◦𝑌
◦𝑌 𝑔 𝑋 𝑋
◦𝑌 𝑔 𝑋, 𝑍 𝑋 𝑍
◦𝑌 𝑔 𝑋, 𝑍

◦ What are the characteristics of new RV 𝑌?


◦ What is the PDF/CDF, 𝑓 𝑦 /𝐹 𝑦 , of the new RV?
Transformation of RVs
◦ For 𝑋~𝑁 𝜇, 𝜎
◦ Let 𝑌 𝑔 𝑋 ; What is the PDF of 𝑌?
Transformation of RVs: Applications
◦ Received Signal at receiver
𝑦 𝑃ℎ𝑥 𝑛
◦ 𝑃: transmit power
◦ ℎ: channel gain (usually complex Gaussian RV)
◦ 𝑛: noise (usually complex Gaussian RV with power 𝑁 )
◦ System performance depends on SNR 𝛾
𝑃ℎ
𝛾
𝑁
◦ Note 𝛾 𝑔 ℎ is a function of RV ℎ
◦ We know PDF of ℎ but need to find CDF of 𝛾 to find
outage probability
Transformation of RVs: Applications
◦ Transmit power might be adaptive eg Cognitve Radio
◦ Received Signal at receiver
𝑦 𝑃ℎ𝑥 𝑛
◦𝑃 : transmit power
◦ 𝑖: interference channel power gain (exponential RV)
◦ ℎ: channel gain (usually complex Gaussian RV)
◦ 𝑛: noise (usually complex Gaussian RV with power 𝑁 )
◦ System performance depends on SNR 𝛾

𝑖
◦ Note 𝛾 𝑔 ℎ, 𝑖 is a function of RV ℎ and 𝑖
◦ We know PDF of ℎ and 𝑖 but need to find CDF of 𝛾 to find
outage probability
Monotonically Increasing Function
o As X increases 𝑌 𝑔 𝑋 also increases

◦ Probability that 𝑌 𝑦 is calculated as


Pr 𝑌 𝑦 Pr 𝑔 𝑋 𝑦
Pr 𝑌 𝑦 Pr 𝑋 𝑔 𝑦
𝐹 𝑦 𝐹 𝑔 𝑦 CDF

◦ Take derivative w.r.t. 𝑦 to get the PDF 𝑥 𝑔 𝑦


𝑑
𝑓 𝑦 𝑓 𝑔 𝑦 𝑔 𝑦
𝑑𝑦
𝑑𝑥 𝑓 𝑥
𝑓 𝑦 𝑓 𝑥
𝑑𝑦 𝑑𝑦 PDF
𝑑𝑥
Monotonically Decreasing Function
o As X increases 𝑌 𝑔 𝑋 decreases

◦ Probability that 𝑌 𝑦 is calculated as


Pr 𝑌 𝑦 Pr 𝑔 𝑋 𝑦
Pr 𝑌 𝑦 Pr 𝑋 𝑔 𝑦
𝐹 𝑦 1 𝐹 𝑔 𝑦 CDF
𝑥 𝑔 𝑦

◦ Take derivative w.r.t. 𝑦 to get the PDF


𝑑
𝑓 𝑦 𝑓 𝑔 𝑦 𝑔 𝑦
𝑑𝑦
𝑑𝑥 𝑓 𝑥
𝑓 𝑦 𝑓 𝑥
𝑑𝑦 𝑑𝑦 PDF
𝑑𝑥
Increasing/decreasing functions

Increasing Function Decreasing Function


𝑓 𝑥 𝑓 𝑥
𝑓 𝑦 𝑓 𝑦
𝑑𝑦 𝑑𝑦
𝑑𝑥 𝑑𝑥

• Note: 0 • Note: 0

◦ Both these cases can be consolidated to yield PDF as


𝑓 𝑥
𝑓 𝑦
𝑑𝑦
𝑑𝑥
Example
◦ 𝑋 is and exponential RV
𝑓 𝑥 2 exp 2𝑥 𝑢 𝑥
◦ Define a transformation 𝑌 𝑋
◦ What is the PDF of 𝑌?

◦ Solution:
General Case

◦ We want to find PDF of 𝑌


◦ Note: 𝑦 𝑌 𝑦 𝑑𝑦 maps to
𝑥 𝑋 𝑥 𝑑𝑥 ∪ 𝑥 𝑋 𝑥 𝑑𝑥 ∪ 𝑥 𝑑𝑥 𝑋 𝑥 ∪ 𝑥 𝑑𝑥 𝑋 𝑥
◦ Pr 𝑦 𝑌 𝑦 𝑑𝑦
Pr 𝑥 𝑋 𝑥 𝑑𝑥 ∪ 𝑥 𝑋 𝑥 𝑑𝑥 ∪ 𝑥 𝑑𝑥 𝑋 𝑥
∪ 𝑥 𝑑𝑥 𝑋 𝑥
◦ Note: 𝑥 are the roots of the equation 𝑦 𝑔 𝑥
General Case
◦ Pr 𝑦 𝑌 𝑦 𝑑𝑦 Pr 𝑥 𝑋 𝑥 𝑑𝑥 Pr 𝑥 𝑋 𝑥 𝑑𝑥
Pr 𝑥 𝑑𝑥 𝑋 𝑥 Pr 𝑥 𝑑𝑥 𝑋 𝑥

◦ For a RV 𝑍: Pr 𝑧 𝑍 𝑧 𝑑𝑧 𝑓 𝑧 𝑑𝑧 as 𝑑𝑧 → 0,
◦ Applying to above expression yields
𝑓 𝑦 𝑑𝑦
𝑓 𝑥 𝑑𝑥 𝑓 𝑥 𝑑𝑥 𝑓 𝑥 𝑑𝑥 𝑓 𝑥 𝑑𝑥
𝑓 𝑥 𝑑𝑥 𝑓 𝑥 𝑑𝑥 𝑓 𝑥 𝑑𝑥 𝑓 𝑥 𝑑𝑥
𝑓 𝑦
𝑑𝑦 𝑑𝑦 𝑑𝑦 𝑑𝑦
◦ In general, this can be expressed as
𝑓 𝑥
𝑓 𝑦
𝑑𝑦
𝑑𝑥
𝑥 are the roots of the equation 𝑦 𝑔 𝑥
Example
◦ Obtain the PDF of RV given that RV
has PDF and CDF
Example
Examples

◦Practice more examples and


questions from book

You might also like