wijh(n+1)=wijh(n)2μ5x˜(n)[ |x˜(n)|2R2 ]f2(j=1Jwjo(n)vjJ(n))×(i=1Iwijh(n)v(ni)+wjd(n)vjJ(n1)+wj,j+1d(n)vj+1J(n1))(4.59) +wj1,jd(n)vj1J(n1) )wjo(n)y(ni)

where μ5 is the weight iteration step size of the hidden layer.

4.4.3Computer simulation results

The input signals are 4PAM and 8PAM. The additive Gauss white noise is zero mean and unit variance. The simulation channels adopt the typical telephone channel and the ordinary channel. Their z-transforms are shown in eqs. (2.84) and (3.13).

When 4PAM and 8PAM signals go through the typical digital telephone channel, the iterative step sizes are μ1 = μ2 = μ3 = μ4 = μ5 = 0.00016 and μ1 = μ2 = μ3 = μ4 = μ5 = 0.0000055, ...

Get Blind Equalization in Neural Networks now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.