An Adaptive Decoding Algorithm Based on CE Criterion for BICM-ID

Bit-interleaved coded modulation with iterative decoding (BICM-ID) can achieve better performance than BICM in both AWGN and Rayleigh fading channels because of the feedback algorithm. However, some existing decoding schemes in BICM-ID cannot perform well in both computational complexity and decoding performance. The Max-Log-MAP algorithm has lower computational complexity with worse decoding performance. Meanwhile, the decoding effect of Log-MAP algorithm is greatly improved but it has much computational complexity. Besides, the Constant-Log-MAP algorithm is between the two algorithms mentioned above which can make a balance in the system´s computational complexity and decoding performance. In this paper, we propose an adaptive decoding algorithm based on cross-entropy (CE) stopping criterion for BICM-ID system. The proposed algorithm employ different decoding algorithm as SNR changes to take advantage of the three algorithms above.