📄 bsep.m
字号:
function [sdb] = bsep(x,y)% bsep - Evaluates the quality of Blind Signal Separation%% usage: sdb = bsep(x,y);%% x(:,i,j) - The i^th microphone signal with only the j^th source active% y(:,i,j) - The i^th separated signal with only the j^th source active% sdb - The attenuation achieved by BSS in dB%% See also: http://www.ele.tue.nl/ica99/% and: "Evaluation of Blind Signal Separation", in proc.% int. workshop on Independent Component Analysis and Blind% Signal Separation"% Author: D.W.E. Schobben (ds@altavista.net)% rev. 1.02 @ 990105if nargin ~=2 | length(size(x)) ~= 3 | length(size(y)) ~= 3, error('Please supply two 3D arrays')endJ = size(x,2);if size(x,3) ~= J | size(y,2) ~= J | size(y,3) ~= J, error('Incorrect array size')endfor j=1:J, ymj = mean(y(:,j,j).^2); ymji = mean( ( sum(y(:,j,:),3) - y(:,j,j) ).^2 ); sdb(j) = 10*log(ymj/ymji);end
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -