radicalfunction
Newbie level 3
Hi,
I'm doing my final year project about blind equalizer based on kernel methods(I'm beginner with Matlab).I have implement this code based on this research "Generalized Correlation Function: Definition,Properties, and Application to Blind equalization".you can find it online
i dont know exactly how to implement gaussian kernel and then the correntropy.
some of the algorthim i usedu can find them in the reaseach paper given above).
the generalized correlation function (correntropy)
V[m] = 1/(N-m+1) sum(k(x-x(n-m)) ,whereas the summation is from n=m to N
or V[m] = E[k(x-x(n-m))]
guassian kernel is given :
k(x-y) = (1/(sigma*sqrt(2*pi)))*exp(||x-y||^2 / (2*sigma^2))
x is 1xN matrix ,N=2000
I'm doing my final year project about blind equalizer based on kernel methods(I'm beginner with Matlab).I have implement this code based on this research "Generalized Correlation Function: Definition,Properties, and Application to Blind equalization".you can find it online
i dont know exactly how to implement gaussian kernel and then the correntropy.
some of the algorthim i usedu can find them in the reaseach paper given above).
the generalized correlation function (correntropy)
V[m] = 1/(N-m+1) sum(k(x-x(n-m)) ,whereas the summation is from n=m to N
or V[m] = E[k(x-x(n-m))]
guassian kernel is given :
k(x-y) = (1/(sigma*sqrt(2*pi)))*exp(||x-y||^2 / (2*sigma^2))
x is 1xN matrix ,N=2000