Filter Design Toolbox

maxstep

Return the maximum step size that allows an adaptive filter to converge

Syntax

• mumax = maxstep(ha,x)
  [mumax,mumaxmse] = maxstep(ha,x)
  

Description

mumax = maxstep(ha,x) predicts a bound on the step size to provide convergence of the mean values of the adaptive filter coefficients. The columns of the matrix X contain individual input signal sequences. The signal set is assumed to have zero mean or nearly so.

[mumax,mumaxmse] = maxstep(ha,x) predicts a bound on the adaptive filter step size to provide convergence of the LMS adaptive filter coefficients in mean square. A warning is issued if ha.stepsize is outside of the range 0 < ha.stepsize < mumaxmse/2.

Note    maxstep is available for the following adaptive filter objects: --adaptfilt.blms --adaptfilt.blmsfft --adaptfilt.lms --adaptfilt.nlms --adaptfilt.se

Examples

Analyze and simulate a 32-coefficient (31st-order) LMS adaptive filter object. To demonstrate the process, run 2000 iterations and 50 trials.

• % Specify [numiterations,numexamples] = size(x);
  x = zeros(2000,50);
  d = x;
  ha  = fir1(31,0.5);  % FIR system to be identified
  for k=1:size(x,2);   % Create input and desired response signal 
                       % matrices
  % Set the (k)th input to the filter
  x(:,k) = filter(sqrt(0.75),[1 -0.5],sign(randn(size(x,1),1))); 
  n = 0.1*randn(size(x,1),1);     % (k)th observation noise signal
  d(:,k) = filter(ha,1,x(:,k))+n; % (k)th desired signal end
  mu = 0.1;                       % LMS step size
  ha = adaptfilt.lms(32,mu);
  [mumax,mumaxmse] = maxstep(ha,x);
  

See Also

adaptfilt/msepred, adaptfilt/msesim, adaptfilt/filter

max

mfilt