Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 1

CHAPTER

NLMS DERIVATION:

e(n)=d(n)-ŵ(n)u(n)

ŵ(n+1)=ŵ(n)+µ̄e*(n)/||u||2 u(n)

ŵ(n+1)=ŵ(n)+½λ*u(n)

d(n)=(ŵ(n)+½λ*u(n))Hu(n)

d(n)=ŵH(n)u(n)+½λ||u(n)||2

where, λ=2e(n)/||u(n)||2

ŵ(n+1)=ŵ(n)+½[ 2e(n)/||u(n)||2] u(n)

ŵ(n+1)=ŵ(n)+[e(n)/||u(n)||2] u(n)

µ̄ can be seen as a positive real scaling factor that control the change of direction in the tap-weights
from one iteration to another.

Final updated weight vector can be expressed as;

ŵ(n+1)=ŵ(n)+[µ̄ /||u(n)||2] e(n)u(n)

You might also like