DSPRelated.com
Forums

phase noise

Started by aamer January 15, 2007
Dear all,

I am modelling phase noise for OFDM systems. At the receiver

r=s*exp(j*phi)+w

r= received signal
s=transmitted signal
phi=random phase
w = AWGN noise

I have studied in the literature that phase noise can be described as
sampled version of wiener process, but for the purpose of analysis and
simulations a discrete time random walk or weiner levy process is used.
What exactly is this weiner levy process, how could i model it in matlab?


aamer wrote:

> I have studied in the literature that phase noise can be described as > sampled version of wiener process, but for the purpose of analysis and > simulations a discrete time random walk or weiner levy process is used. > What exactly is this weiner levy process, how could i model it in matlab?
The Wiener process is a special case of the Levy process. For your purposes, it's probably sufficient to generate a Wiener process by: x = randn(N.1); h = some filter with the appropriate Fourier transform y = conv(x,h); Basically generate a discrete-time white Gaussian sequence and then filter it using filter h. Choose h according to what you want. I have seen phase noise models where h is a lowpass filter, for example. Hope that helps, Julius