Research On Theory And Technology For Wireless Network Coding

With the soar development of wireless communication technology, an increasing number of users are sharing more and more abundant wireless information services. This makes the limited wireless resources face a huge challenge. Relay technology makes the wireless resources be further used efficiently, and introducing network coding technology can improve the spectral efficiency of wireless relay network. The traditional network coding can ensure the validity and reliability only with a very low error rate. However, the complexity of the wireless channel is always resulting in a high bit error rate. Thus, joining channel coding and network coding can really improve the performance of the wireless system. It can obtain diversity gain, channel coding gain and network coding gain simultaneously when designing joint channel-network coding system based on wireless relay network.This thesis studied the design of network coding system combining channel coding based on the transmission scenarios of multiple access wireless relay network. Two kinds of new network coding methods were designed, each of which was innovation of this thesis. First, to avoid the phase ambiguity of coherent demodulation, this thesis employed differential modulation to design network coding scheme, and used the idea of bit interleaved coded modulation with iterative decoding to make up for the lost of performance which was brought by differential modulation. It researched two channel codes (convolution code and low density parity check code) matching differential quadrature amplitude modulation under gray and modified set partitioning mappings. Second, based on q-ary low density parity coding, a kind of q-ary network coding method in Galois field was proposed. It studied this new method in a large-scale wireless scenario with multiple users and multiple relays. Owing to the sparse parity relationship between users and relays, this thesis employed the rule of check node updating to process network decoding, and introduced iterative mechanism between channel decoding and network decoding. This method made full use of multiuser diversity, network diversity and coding gain. Moreover, this thesis designed the corresponding binary method. The simulation results showed that q-ary method could achieve significant performance gain compared with the binary method. In all, this thesis studied the joint channel-network coding technology deeply and promoted reliable and efficient network coding methods.