Bind deconvolution via maximum cyclostationarity
Matlab documentation
M. Buzzoni, J. Antoni and G. D’Elia
Last up date: December 2017
Folder contents and file descriptions
MaxCycloBD.m
This function performs the blind deconvolution of signal x for a
single-input-single-output system by finding the optimal inverse
filter (FIR form) that maximizes the cyclostationarity of the out-
put.
MaxCycloBD_SIMO.m
This function performs the simultaneous blind deconvolution of
multiple signals x for a single-input-multiple-output system by
finding the optimal inverse filters (FIR form) that maximizes the
cyclostationarity of the output.
MaxCycloBDangle.m
This function performs the blind deconvolution of signal x for a
single-input-single-output system by finding the optimal inverse
filter (FIR form) that maximizes the cyclostationarity of the out-
put through a weighting matrix defined in the angle time/angle
domain.
demo_CYCBD.m
This interactive script shows six different applications of the pre-
vious functions, extracting cyclostationary sources from noisy ob-
served signals and considering different interferences.
Syntax
% [h,s,kappa,count,err] = MaxCycloBD(x,N,alpha,fs,param,p)
% SISO BLIND DECONVOLUTION VIA MAXIMUM CYCLOSTATIONARITY
% Blind deconvolution of signal x by finding the optimal inverse filter (FIR form)
% that maximizes the cyclostationarity of the output.
%
%--------
% Inputs
%--------
%
% x.........observed signal
% N.........FIR filter length
% alpha.....cyclic frequency set (in the form of a vector)
% fs........sampling frequency of x
% param.....structure of setting parameters organized as follows:
% param.ER......minimal relative error on result (default value = 1e-3)
% param.iter....maximum number of iterations (default value = 50)
% p.........cyclostationarity order to maximize (default = 2)
%
1
