chan_sim.m

来自「《Wireless Communications and Networking》」· M 代码 · 共 66 行

%The function rayleigh returns a correlated rayleigh random variates by Inverse Discrete
%Fourier Transform.  
%Inputs : fm   - the maximum Doppler frequency normalized by the sampling rate
%         N    - the number of samples
%         std  - standard deviation for the two zero mean Gaussian variates.
%Outputs : magnitude  - Magnitude of the generated baseband rayleigh fading sequence 
%          theta      - Phase of the generated baseband rayleigh fading sequence
%          corr       - Correlation of the
%          var_r      - variance of the rayleigh fading sequence.

function [magnitude, theta, corr, var_r] = chan_sim (fm, N, std)

A = std*randn(1, N);
B = std*randn(1, N);

km = floor(fm*N);

Fk(1) = 0;
for i = 2:km
  Fk(i) = sqrt(1/(2*sqrt(1-((i-1)/(N*fm))^2)));
end

Fk(km+1) = sqrt(km/2*(pi/2-atan((km-1)/sqrt(2*km-1))));
for i = km+2:N-km
  Fk(i) = 0;
end

Fk(N-km+1) = sqrt(km/2*(pi/2-atan((km-1)/sqrt(2*km-1))));
for i = N-km+2:N
  Fk(i) = sqrt(1/(2*sqrt(1-((N-(i-1))/(N*fm))^2)));
end
temp = 0;
for k = 1:N
   Z(k) = A(k).*Fk(k) - j*B(k).*Fk(k);
   temp = temp + (Fk(k)/N)^2;
end
   
var_r = 1/2;

Zn = ifft(Z, N)/std/sqrt(2*temp);
for k = 1:N
   magnitude(k) = abs(Zn(k));
   theta (k) = angle(Zn(k));
end

%Calculate correlation.  Note correlation is calculated for delta_t of 2Ts.
for k = 0:100
   for j = 1:N
      magnitude2(j) = magnitude(mod(j+k*2,N)+1);
   end
   temp(1:N) = magnitude(1:N).*magnitude2(1:N);

   mean1 = mean(magnitude(1:N));
   mean2 = mean(magnitude2(1:N));
   mtemp = mean(temp(1:N));

   cov = mtemp - mean1 * mean2;

   var1 = mean(magnitude(1:N).*magnitude(1:N))-mean1^2;
   var2 = mean(magnitude2(1:N).*magnitude2(1:N))-mean2^2;
   
   corr(k+1) = cov/sqrt(var1*var2);
end

return;