where μ3 and μ4 are the expansion factor and the translation factor, respectively.

The iteration step-size factors are

J(n)a(n)=2[ | x˜(n) |2R2 ]| x˜(n) |2| x˜(n) |a(n)(7.50)

J(n)b(n)=2[ | x˜(n) |2R2 ]| x˜(n) |2| x˜(n) |b(n)(7.51)

a(n)=aR(n)+jaI(n)(7.52)

b(n)=bR(n)+jb1(n)(7.53)

| x˜(n) |a(n)=| x˜(n) |aR(n)+j| x˜(n) |aI(n)(7.54)

| x˜(n) |b(n)=| x˜(n) |bR(n)+j| x˜(n) |bI(n)(7.55)

| x˜(n) |aR(n)=x˜(n)x˜*(n)aR(n)=12| x˜(n) |x˜(n)x˜*(n)aR(n)=12| x˜(n) |(f2(vR(n))+f2(vI(n)))aR(n)=1| x˜(n) |[ f(vR(n))

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.