Hi all,
I hope you would be able to help me out. I have written a turbo code encoder and decoder in MATLAB with the specs below:
Channel Additive White Gaussian Noise (AWGN)
Modulation Binary Phase Shift Keying (BPSK)
Component
Encoders -
Two identical recursive
convolutional codes
RSC
Parameters
n = 2, k = 1,K = 3
G0 = 7, G1 = 5
Interleaver
1000-bit random interleaver
Puncturing
Used
Half parity bits from each component
encoder transmitted give half-rate code
Component
Decoders
Log-MAP
decoder
Iterations 8
I expected for 8 iterations, that the BER would be very small for high snr, but to my surprise, This is not so. the BER is still high in comparison to other journals results. The alg is correct to my knowledge. Thanks in advance for your help.