How to save a loop data

Hi guys, I have a for loop, but every iteration overwrites the variable(max_run), and I have only the final data left. How can I save data from every loop? I saw some other questions like my issue, but I always get an error"Improper assignment with

%%BPSK Generation
clear all; close all; clc;
N = 2*10^4;                        % number of bits or symbols
ip = rand(1,N)>0.5;                % generating 0,1 with equal probability
s = 2*ip-1;                        % BPSK modulation 0 -> -1; 1 -> 1

%%Prepare transmitted signal

m=2;                               % oversampling factor
beta=0.3;                          % rolloff parameter for SRRC
l=100;                             % 1/2 length of pulse shape (in symbols)
chan=[1];                          % T/m "channel"
toffset=-0.5;                      % initial timing offset
pulshap=srrc(l,beta,m,toffset);    % SRRC pulse shape
sup=zeros(1,N*m);                  % upsample the data by placing... 
sup(1:m:end)=s;                    % ... p zeros between each data point
hh=conv(pulshap,sup);              % ... and pulse shape
r1=conv(hh,chan);                  % ... to get received signal

%%Noise Generation

SNRdB=1:2:13;                            % Signal to Noise Ratio
SNR=10.^(SNRdB/10);

for cv=1:length(SNR);

no=sqrt((1/SNR(cv)))*randn(1,40400); % Noise generation
    r=r1+no;                             % Adding the noise to the received signal

      %%Matched Filter & Convolution process

      matchfilt=srrc(l,beta,m,0);          % filter = pulse shape
      x=conv(r,matchfilt);                 % convolve signal with matched filter

      %%Run clock recovery algorithm

      tnow=2*l*m+1; tau=0; xs=zeros(1,N);         % initialize variables
      tausave=zeros(1,N); tausave(1)=tau; i=0;
      mu=0.05;                                    % algorithm stepsize
      delta=0.1;                                  % time for derivative

      while tnow<length(x)-2*l*m                  % run iteration
        i=i+1;
        xs(i)=interpsinc(x,tnow+tau,l);           % interpolated value at tnow+tau
        x_deltap=interpsinc(x,tnow+tau+delta,l);  % get value to the right
        x_deltam=interpsinc(x,tnow+tau-delta,l);  % get value to the left
        dx=x_deltap-x_deltam;                     % calculate numerical derivative  
        tau=tau+mu*dx*xs(i);                      % alg update (energy)
        tnow=tnow+m; tausave(i)=tau;              % save for plotting
        indexmin = find(tausave >= 0.5);          % finding iteration value.. 
        max_run = min(indexmin)                        % automaticlly 
      end

      % Plot results
      figure, subplot(2,1,1),  plot(xs(1:i-2),'b.'); 
      legend([ 'SNR=' int2str(SNR(cv))]);        % plot constellation diagram
      title('constellation diagram');
      ylabel('estimated symbol values')
      subplot(2,1,2), plot(tausave(1:i-2)) 
      ylabel('offset estimates'), xlabel('iterations')
      legend([ 'SNR=' int2str(SNR(cv))]);

end

[Walter Roberson]

Are you getting srrc() from http://www.gaussianwaves.com/2011/04/square-root-raised-cosine-filter-matchedsplit-filter-implementation-2/ ?

Answer 1

[Walter Roberson]

max_run(i) = min(indexmin);

[Bahaa Algebory]

Thanks for you Walter, But I got the same error "Improper assignment with rectangular empty matrix", and there is a red line appears under 'max_run' saying "The variable 'max_run' appears to change size on every loop iteration". Actually, there should be 7 values for 'max_run' according to SNR loop, but I'm getting just the last value.

[Walter Roberson]

Can you guarantee in

indexmin = find(tausave >= 0.5);          % finding iteration value..

that the find() will find something ? If not then you could be taking min() of an empty vector, which is going to have an empty result which you would then try to save.

[Bahaa Algebory]

Walter, thank you again,

Yes, there are values for 'indexmin'. This is the code of 'srrc' function.

function s=srrc(syms, beta, P, t_off);

% s=srrc(syms, beta, P, t_off);
% Generate a Square-Root Raised Cosine Pulse
%      'syms' is 1/2 the length of srrc pulse in symbol durations
%      'beta' is the rolloff factor: beta=0 gives the sinc function
%      'P' is the oversampling factor
%      't_off' is the phase (or timing) offset 

if nargin==3, t_off=0; end;                       % if unspecified, offset is 0
k=-syms*P+1e-8+t_off:syms*P+1e-8+t_off;           % sampling indices as a multiple of T/P
if (beta==0), beta=1e-8; end;                     % numerical problems if beta=0
s=4*beta/sqrt(P)*(cos((1+beta)*pi*k/P)+...        % calculation of srrc pulse
     sin((1-beta)*pi*k/P)./(4*beta*k/P))./...
   (pi*(1-16*(beta*k/P).^2));

Actually, now I'm required to find the value of offset point" Y axis" where the curve becomes stable for the same value which is about "0.5" by depending on iteration values. So, any help will be appreciated.

[Walter Roberson]

Your line

          xs(i)=interpsinc(x,tnow+tau,l);           % interpolated value at tnow+tau

uses an undefined routine interpsinc . I grabbed interpsinc from the File Exchange but you appear to be asking to interpolate all the x values at time tnow+tau and you have over 40000 x values so the result is over 40000 results, which do not fit into the single location xs(i)