Hi everyone. I am simulating a flat one tap rayleigh channel with ofdm, with the following parameters:
N=1024 number of subcarriers
Ncp=32 cyclic prefix
max_ITER=1000 maximum number of iterations (number of OFDM symbols)-> used in a loop 1:1:max_ITER
max_SNR=20dB maximum number of SNR-> used in a loop 1:1:max_SNR
i know that the output signal should be y=h*x+n
where n is the gaussian noise applied, x the transmitted OFDM symbol and h the rayleigh factor(rayleigh channel). Considering that the number of taps is 1 and that h should change for every ofdm symbol, how am i supposed to choose the factor h?
hello anta,
I am also working on this, can u help me clear some doubts.
1. the o/p Y =conv(h, x)
in that case to decide the lenght of the cyclic prefix u need to know the length of the channel i.e no. of samples in the impulse response of the channel.
I mean if length(x)=M
length(h)=L, then Length=M+L-1.
and so length of CP=L-1.
so how did u arrive at 32??
pls rply..
hello anta,
I am also working on this, can u help me clear some doubts.
1. the o/p Y =conv(h, x)
in that case to decide the lenght of the cyclic prefix u need to know the length of the channel i.e no. of samples in the impulse response of the channel.
I mean if length(x)=M
length(h)=L, then Length=M+L-1.
and so length of CP=L-1.
so how did u arrive at 32??
pls rply..
the channel which i am simulating is flat hence one tap rayleigh channel. In this case convolution becomes a simple multiplication. Since we have only one tap h is a single number which must change for every ofdm symbol...
I use cp as a parameter, which i change to see the simulation results.
The length of frequency response ->v=length(h) should be greater than the length of cp to avoid ISI...This happens when the channel has more than one taps...In both cases though i give cp a certain value. For instance if v=length(h)=10 then to avoid ISI i choose cp to be a num greater than 10->cp=32, otherwise we have ISI->cp=8. Hope this helps