c7_jakes.m

一个不错的jakes模型仿真,是老外写的,取自<<基于MATLAB的无线通信仿真>>一书

% File : c7_Jakes.m
% Generate and test the impulse response of the filter
clear
clc
fd = 100;       % maximum doppler
impw = jakes_filter(fd);    % call to Jakes filter
fs = 16*fd; ts = 1/fs;      % sampling frequency and time
time = [1*ts:ts:128*ts];    % time vector for plot
subplot(2,1,1)
stem(time,impw,'.'); grid;
xlabel('Time'); ylabel('Impulse Response')
% Square the fft and check the power transfer function
[h f] = linear_fft(impw,128,ts);    % generate H(f) for filter
subplot(2,1,2)
plot(f,abs(h.*h)); grid;
xlabel('Frequency'); ylabel('PSD')
% Put Gaussian noise through and check the output psd
x = randn(1,1024);      % generate Gaussian input
y = filter(impw,1,x);   % filter Gaussian input
[output_psd ff] = log_psd(y,1024,ts);   % log of PSD
figure;
subplot(2,1,1)
plot(ff,output_psd); grid;
axis([-500 500 -50 0])
xlabel('Frequency'); ylabel('PSD')
% Filter complex noise and look at the envelope fading
z = randn(1,1024)+i*randn(1,1024);  % generate complex noise
zz = filter(impw,1,z);  % filter complex noise
time = (0.0:ts:1024*ts);    % new time axis
% Normalize output and plot envelope
zz = zz/max(max(abs(zz)));  % normalize to one
subplot(2,1,2)
plot(time(161:480),10*log10(abs(zz(161:480)))); grid;   
axis([0.1 0.3 -20 0])
xlabel('Time'); ylabel('Log Amplitude')