Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Gaussain Noise in Matlab 1/sqrt(2) (randn(no of bits)+j.*(no of bits))

Status
Not open for further replies.

aliazmat306

Newbie level 4
Newbie level 4
Joined
May 28, 2013
Messages
6
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,283
Activity points
1,326
could somebody please brief me the following method of generating Gaussian Noise specially the term sqrt(2), i.e., why it has been used here

Gaussain Noise in Matlab 1/sqrt(2) (randn(no of bits)+j.*(no of bits))

Thank you very much for your time!
 

Hi
It is just to keep variance of each sample to be one. randn() function in MATLAB results in a Gaussian noise with zero mean and unit variance. So, variance of randn() + j randn() would be 2 and to keep its variance to be 1 a scaling factor of 1/sqrt(2) is multiplied.

Hope this helps
 
but simple algebra shows that

1/sqrt(2)[1+1] = 1/sqrt(2)

I mean the variance of sum of two independent normal Gaussian Random variable is 2. (Agree)
but when scalar factor (1/sqrt(2)) is multiplied the results is 1/sqrt(2) not 1.
plz explain, I am confused.
 

Hi
Kindly note that \[var(aX) = a^2 var(X) \]. So, multiplying by \[\frac{1}{\sqrt{2}}\] scales variance by \[\frac{1}{2}\]. Thus total variance is (1+1)/2=1.
 
Thank you very much dear MHanif. This answer is what I wanted.
Thanks Again!!!
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top