Research On Key Technology For Joint Channel And Network Coding

In order to make use of the limited wireless resources, relaying technique is introduced in the next generation mobile communication system which be used to enlarge the cover region and enhance the cell edge quality of service. But the employment of relaying leads to a decline in spectrum efficiency, furthermore, causing overall performance decline of wireless network. Recently, the idea of network coding has been proposed that provides an efficient way to improve the spectrum efficiency of wireless relay network under the constraint of limited resources, which is to allow combining several data packets and then forwarding the aggregation data packets at the relaying nodes, in exchange for the improvement of spectrum efficiency. In this dissertation, various joint channel and network coding approaches are deep studied for wireless multiple access relay channel system. The aim of joint coding scheme is to improve network reliability under the premise of insuring transmission efficiency.The main contributions of this dissertation as follows:From the Turbo en/decoding theory, and combining the idea of network coding, an joint Turbo and network coding scheme is proposed which based on serial relay. For the network coding scheme with the help of one half-duplex relay, only the extra redundancy data packets of two sources’XOR-operation messages are forwarded by the relay node. The corresponding iterative decoding strategy is designed at the base station (BS), meanwhile Turbo iterative decoding is constructed to each source node which is used to improve the reliability of network transmission. Moreover, in order to further improve the transmission efficiency, joint Turbo and network coding scheme is proposed which under the help of parallel relay, the relay node is allowed to receive the data information from source node and forward the network coding information to the destination node at the same time. Meanwhile using the principle of Turbo en/decoding, the bit interleaver and soft-input soft-output decoder are introduced to the proposed scheme, combining the LogMAP algorithm and the idea of iterative decoding to achieve decoding which is used to enhance the quality of network transmission. Research shows that the proposed network coding schemes outperform the simple decode and forward of relay respectively.From the view of encoding, and belief propagation algorithm which considered as iterative decoding algorithm about low-density parity-check (LDPC) codes, which combining" the theory of product codes, the dissertation proposes a network coding scheme based on LDPC product codes. In the proposed scheme, the relay node employs product coding instead of XOR-operation to combine the data packets from the two sources and to generate extra redundancy. At the BS, the extra redundancy from the relay node together with the data packets received from two source nodes forms a complete product coding matrix in which each row and each column are independently coded, and designs the iterative decoding algorithm of LDPC product codes which is used to obtain the original information of each source node. Furthermore, the optimal algorithm of extrinsic information transfer (EXIT) chart together with Gaussian increment is proposed which is used for optimal degree distribution of LDPC product codes, meanwhile, the convergence behavior of iterative decoding for LDPC product codes is predicted. Numerical results show that the proposed scheme substantially improves the bit-error-rate (BER) performance as compared with the traditional network coding scheme; Moreover, the optimized scheme also improves the BER performance, meanwhile, the EXIT charts are used to approximately predict the performance of iterative decoding.From the hierarchical transmission principle of radio broadcast, and combining the joint information processing of relay node, the dissertation proposes a network coding scheme based on superposition codes. At each source node, hierarchical modulation based on superposition codes is introduced, which broadcasts the data information to the relay node and BS by using the hierarchical modulated transmission scheme. According to the different channel quality of relay link and direct link, the relay node re-encodes the error-free data information received from both sources, and then forwards the coded information to the BS. The BS only receives the high-priority data information through direct link, using the high-priority data information and the data information which is forwarded by relay node to design the decoding scheme, and further to recover the original data information of each source node. Simulation results indicate that the proposed scheme improves the BER, as compared with the traditional network coding scheme, while the gap between direct and relay link is enough.Based on multi-source (greater than2) single-relay transmission scenario, considering the direct links, this dissertation proposes one kind of multi-source relay network coding approaches; Without considering the direct links, this dissertation derive the mathematical expression of multi-source relay network coding matrix. In the multi-source single-relay network coding transmission scheme with considering direct links, the expression of multi-source coding matrix has been provided at the relay node, the implementation process of iterative decoding algorithm also has been provided at the BS, meanwhile, the EXIT charts are used to analyze the performance of iterative decoding scheme in details. The EXIT analysis and simulation results show that the proposed scheme has slightly poor BER performance than the maximum ratio combining (MRC) scheme at low signal to noise ratio (SNR) regime. However, once the SNR level is slightly high for the proposed scheme which can achieve better performance gain as compared with the referenced scheme. Furthermore, following the increase of channel quality for relay link, the BER performance of proposed scheme also has been improved obviously.