An Efficient Soft-Input Soft-Output Decoder for Polar Codes in MIMO Iterative Detection System

Abstract

Iterative detection and decoding (IDD) induces higher performance than separate detection and decoding schemes. However, IDD requires the exchange of soft information between the detector and decoder, thus decoders should be able to process soft input and provide soft output (SISO). Existing SISO polar decoders such as belief propagation (BP) and soft cancellation (SCAN) give poor block error rate (BLER) performance compared with cyclic redundancy check (CRC)-aided successive cancellation list (CA-SCL) decoders, whereas other more sophisticated SISO decoders have high complexity albeit good BLER performance. In this paper, an efficient SISO decoder that provides both high BLER performance and low complexity is proposed for polar codes in IDD system. Specifically, the proposed SISO decoder employs the sum-product algorithm over the sparse parity-check matrix representation of polar codes for lower complexity and uses the best candidates from list decoders for good error correction performance. We comprehensively evaluate the proposed SISO decoder performance in MIMO iterative system. Simulation results show that the proposed SISO decoder outperforms the existing SCAN and BP decoders and achieves similar BLER performance, but lower complexity compared with the state-of-the-art Soft List decoder.