DSPRelated.com
Forums

OFDM: Mapping of Rayleigh SNR to AWGN SNR for BLER calculation

Started by Arthur August 29, 2005
Dear all,

Hi, I am trying to simulate the performance (block error rate) of ofdm
over raylegih. My question is, how do I calculate SNR for OFDM Rayleigh
channel and map it to equivalent AWGN SNR so that I can calculate the
block error rate ? I like to mention that I have already simulate AWGN
block error rate (BLER) for different modulation and coding (MCS)
combination.

Difficulty arises due to the fact that there are many subcarriers and
each on of them have different SNR in rayleigh channel. Taking the mean
of Rayleigh SNR doesn't sound like a brilliant way ....

Thanks in advance !!

Regards,
Arthur.

On 29 Aug 2005 17:16:55 -0700, "Arthur" <geniusbb@gmail.com> wrote:

>Dear all, > >Hi, I am trying to simulate the performance (block error rate) of ofdm >over raylegih. My question is, how do I calculate SNR for OFDM Rayleigh >channel and map it to equivalent AWGN SNR so that I can calculate the >block error rate ? I like to mention that I have already simulate AWGN >block error rate (BLER) for different modulation and coding (MCS) >combination. > >Difficulty arises due to the fact that there are many subcarriers and >each on of them have different SNR in rayleigh channel. Taking the mean >of Rayleigh SNR doesn't sound like a brilliant way ....
Uh, not quite sure what you're asking here. Are you trying to find a way to estimate AWGN performance from performance in Rayleigh statistics? Vice versa? If so, I wouldn't rely on the results too heavily. Usually people do sims in Rayleigh statistics to find out things that the AWGN results may not show. There may not be a reliable way to map from one to the other (I'd expect that there isn't in general). I'm also not sure whether you're just using Rayleigh distributed noise statistics or whether you're applying a fading channel model with Rayleigh statistics. Those are different ideas as well. Eric Jacobsen Minister of Algorithms, Intel Corp. My opinions may not be Intel's opinions. http://www.ericjacobsen.org