On Distributed Joint Source-channel Coding

Distributed joint source-channel coding is one of optimized coding schemes, which can not only make full use of the characteristics of distributed sources, but also the advantages and disadvantages of source coding and channel coding. So distributed joint source-channel coding has become a hot research spot since it has been put forward. This thesis is mainly concerned with the design of joint source-channel coding schemes, and the author’s main work is outlined as follows:First, an optimized scheme of distributed joint source-channel coding is studied. For the distributed joint source-channel coding model, we have analyzed the distributed source coding scheme based on syndrome former, and carried out performance simulations under different channel conditions. Then, an optimized design scheme of distributed joint source-channel coding for the distributed source coding based on syndrome former is proposed. Under the same conditions for the related and actual channels, simulations are carried out for the proposed scheme. Simulation results show that the optimized scheme is superior to the distributed source coding scheme based on syndrome former.Second, a distributed joint source-channel coding scheme based on superposition coding is studied. For the non-symmetric channel with the better channel is unavailable, a superposition coding based distributed joint source-channel coding scheme is presented, in which the allocation and optimization of modulation signal power are used to achieve the best performance under the non symmetric channel, and the system capacity of this scheme is analyzed. The superposition coding scheme under the BPSK modulation through an asymmetric AWGN channel and the joint decoding strategy are described in detail. Furthermore, a comparison is made between the independent decoding strategy and the joint decoding strategy. Compared with conventional distributed joint source-channel coding scheme, the proposed scheme has the same algorithm complexity, and can achieve better performance by matching two asymmetric channels and taking full advantage of the better channel. So this is a more practical distributed joint source-channel coding scheme.