The problem of representing signal in complex form for turbo decode

Hi all,

At the moment, I have a problem in solving turbo decoding when I represent the received signal in complex form.

Assume that the received signal is as (same form for systematic and parity symbols, BPSK is used): y=sqrt(SNR)*h*x+n, in which:

+) h is channel coefficients, h=1/sqrt(2)*(randn(1)+1i*rand(1)), channel is flat fading, h is constant for a block of N size

+) x is a sequence of transmitted symbol

+) n is Gaussian noise, n=1/sqrt(2)*(randn(1,N)+1i*rand(1,N))

+) N is the length of sequence

The received signal work as the input of turbo decoder then, but the received signal here is complex. What I need to do with it ?

Thanks for the help