clear; clc; data_size=100; M=2; freq_sep=500; nsamp=6; fs=10000; n=2;
y=wavread('voice');
partition=[-.8,-.6,-.5,-.4,-.3,-.2,-.1,0,.1,.2,.3,.4,.5,.6,.8];
codebook=[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15];
% Now optimize, using codebook as an initial guess. [partition2,codebook2] = lloyds(y,codebook); [index,quants,distor] = quantiz(y,partition,codebook); [index2,quant2,distor2] = quantiz(y,partition2,codebook2); %Compare mean square distortions from initial and optimized [distor, distor2] % parameters. quantization=dec2bin(quants,4);
*x=str2num(quantization);
fsk_mod=fskmod(x,M,freq_sep,nsamp,fs); *noise=awgn(fsk_mod,0); fsk_demod=fskdemod(noise,M,freq_sep,nsamp,fs); [num,rate]=biterr(quants,fsk_demod) sound(fsk_demod)
[001;011;111;111] is not binary numbers: it is a column array of decimal numbers. one. eleven. one hundred and eleven.
If instead your matrix is A = [0 0 1;0 1 1; 1 1 1; 1 1 1] then you want
t = A.'; B = t(:) .';
sir
I am implementing bfsk for an audio file of .wav format. For bfsk input to fskmod should be binary. I am having problem in quantizing the sampled audio file. If you please read the code and suggest something?
dec2bin() does not produce binary: it produces strings of '0' and '1'
0 Comments