Distributed Multiple-Component Turbo Codes for Cooperative Hybrid ARQ
Bo Zhang, Hong Chen, Mohammed El‐Hajjar, Robert G. Maunder, Lajos Hanzo · IEEE Signal Processing Letters · 2013
We design distributed multiple-component turbo-codes (MCTCs) for cooperative automatic repeat request (CARQ) aided systems, which are capable of attaining both a similar performance and throughput as the conventional twin-component turbo code (TCTC) aided system, while imposing a significantly lower decoding complexity. The reduction in the decoding complexity of the MCTC-CARQ scheme is a direct benefit of using low-memory MCTC component encoders, combined with the novel principle of deferred iteration (DI), where the decoding process is only initiated when sufficient information has been accumulated at the receiver so that an open tunnel exists in the Extrinsic information transfer (EXIT) charts. In the presence of multiple collaborating relay nodes (RNs), the specific RN which succeeds in decoding the source message, and additionally, has the highest signal-to-noise ratio at the destination node is selected as the retransmission node. Our numerical results demonstrate that the proposed MCTC-CARQ protocol is capable of reducing the decoding complexity by up to 40% compared to its TCTC-CARQ counterpart, while achieving the same throughput and reliability.