How to find autocorrelation function for a mm-sequence ?

1 次查看(过去 30 天)
I have generated an m-sequence with periodicity 1023 i.e., for order 10. I have successfully generated the m sequence using matlab but got struck while evaluating autocorrelation function of that sequence. I tried using 'xcorr' but it doesn't seem to match the text book i.e., when I use xcorr I get values more than 1. The problem persist even if I try to normalize it i.e., when I divide it by 1023.
If anyone can help me out how to evaluate the autocorrelation function it will be helpful to me. Also, I am wondering if I could evaluate the autocorrelation for specified time value. For example I know matlab calculates the autocorrealtion for discrete values 1,2,3 ... I was wondering if I could calculate it for 0.2,0.4.
Thanks guys in advance.

采纳的回答

Greg Heath
Greg Heath 2013-4-11
I do not have xcorr. However, I believe there are at least four options with names like 'unbiased', 'biased' , 'coeff' and 'none' see
help xcorr doc xcorr
for details. Also check to see if you have crosscorr
help crosscorr doc crosscorr
Finally, you can use
crosscorrelation(a,b) = ifft(conj(fft(a)).*fft(b))/N;
Unfortunately, nncorr in the Neural Network Toolbox is buggy.
Hope this helps.
Thank you for formally accepting my answer
Greg

更多回答(0 个)

产品

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by