Study On Cooperative Decoding Of The LDPC-based Relay System

The combination of transmission scheme of relay system and channel coding/decoding technology has become one of the research focuses in modern communication systems.Low-density Parity-check(LDPC)code,one of the capacity-approaching codes,has been widely used to different communication environments.Cooperative communication technology of the relay system can significantly eliminate the error propagation and reduce the decoding complexity.Besides,it can improve the error correction performance by the cooperation and iterative decoding among source node,relay node and destination node.The research of relay system involves multi-user information theory,cooperative communication and coding modulation theory etc.One important direction is to design coding/decoding scheme to improve the system performance through cooperation.At present,the coding/decoding research of relay system mainly focuses on the capacity approximation designing and performance optimization at the expense of high complexity.Coding optimization and signal processing are generally adopted to the design of cooperative communication.This paper researches the decoding strategy and signal modulation scheme of LDPC-based relay cooperative communication system.The completed work and innovation are combined and summarized below:1.Different from the traditional computing methods of extrinsic information,this paper introduces a class of extrinsic information function vector based on hard decision sequence.This vector indicates the differences between two decoders of the destination node.Combining with the new extrinsic information function and the high correlation between the signal from the source-destination link(S-D)and the signal from the relay-destination link(R-D),the decoding scheme of relay system can be conveniently designed.2.An iterative detection/decoding algorithm of correlated sources for the LDPC-based relay systems is presented.For the presented algorithm,the codeword at time t-1 is formulated as a highly correlated counterpart from the codeword at time t.The extrinsic information exchanged between the two decoders is derived by the log-likelihood ratio(LLR)associated with the hard decision XOR vectorz.Simulation results show that the presented algorithm has a slightly better performance than the traditional turbo-like algorithm.Furthermore,the presented algorithm requires fewer computing operations per iteration and has faster(about two times)convergence rate,which is more convenient for hardware implementation and actual application.3.For the relay system of three nodes,a constellation cooperative modulation scheme between source node and relay node is presented.In the scheme,the signal constellation of source node is designed according to the code of previous time slot(corresponding to the decoding results of relay nodes),which achieves beneficial superposition of the constellation signal in destination node and thus can improve system performance.Simulations show that with the optimal power allocation parameter,the iterative detection/decoding algorithm of correlated sources with constellation modulation has a better performance(about 1.2dB)than the original algorithm.