Multi-Level Turbo Decoding Assisted Soft Combining Aided Hybrid ARQ

Hybrid Automatic Repeat reQuest (ARQ) plays an essential role in error control. Combining the incorrectly received packet replicas in hybrid ARQ has been shown to reduce the resultant error probability, while improving the achievable throughput. Hence, in this contribution, multi-level turbo codes have been amalgamated both with hybrid ARQ and efficient soft combining techniques for taking into account the Log-Likelihood Ratios (LLRs) of retransmitted packet replicas. In this paper, we present a soft combining aided hybrid ARQ scheme based on multi-level turbo codes, which avoid the capacity loss of the twin-level turbo codes that are typically employed in hybrid ARQ schemes. More specifically, the proposed receiver dynamically appends an additional parallel concatenated Bahl, Cocke, Jelinek and Raviv (BCJR) algorithm based decoder in order to fully exploit each retransmission, thereby forming a multi-level turbo decoder. Therefore, all the extrinsic information acquired during the previous BCJR operations will be used as a priori information by the additional BCJR decoders, whilst their soft output iteratively enhances the a posteriori information generated by the previous decoding stages. We also present link level Packet Loss Ratio (PLR) and throughput results, which demonstrate that our scheme outperforms some of the previously proposed benchmarks.