Research On Compressive Sensing And Coding Technique In Cooperative Schemes

Cooperative diversity could improve the information throughput rate in the networks and it has become a research hotspot in recent years. For cooperative communication system, correlatively theoretical research has been done methodically and many achievements have been achieved on practical implementations. Source channel coding is one of the most important parts in communication systems, and it is important for the design of cooperative communication system. Designing effective coding scheme for cooperative communication system could achieve great gain. On the one hand, in numerous cooperative schemes, quantize and forward scheme attracts much attention for low complexity and high-performance, but there is no quantize and forward scheme using multi-relay joint coding until now. On the other hand, Compressive Sensing is a new technique for sampling and source coding. In some cases, Compressive Sensing can sample the source message in a much lower rate than Nyquist sampling. Recently, researchers have found that compressive sensing technique can be used for channel coding or joint source channel coding. The study of compressive sensing technique for cooperative schemes is taking off.This paper proposes two kinds of cooperative communication schemes using quantize and forward and compressive sensing technique. In the first scheme, two relays forward the received message cooperatively. Each relay only processes parts of the quantized codeword. The principle of selecting the forwarding part is based on the channel condition. The relay with better channel condition forwards the parts which is more important to decode the source message at the destination. We deduce the calculation of LLR in order to take into account of the bits from different relays. The relays can transmit their own message without affecting the transmission of the source. This paper analyzes the optimization problem of the situation at the relays. The simulation result shows that the designed two relay quantize and forward system utilizes the relays with different channel condition effectively. Compared with traditional quantize and forward system, our system has better performance. Optimal power allocation ratio takes performance improvement. Simulation results verify the effectiveness of optimal power alloction and adaptation in variational envirment through the simulation of real svstem.Then, we propose multi-relay compressive cooperative scheme in which the source and relays utilize compressive sensing as joint source channel coding, and the relays use decode and forward strategy. The scheme considers two cooperative modes, the difference of which is whether the system has relay selection. Based on the two modes, we research four decode and forward strategies. The achievable rates of all strategies have been derived in this paper. For the compressive cooperation system, the simulation results show that the channel conditions of source-to-relay links deside which mode is better. It deponds on the channel conditions that which strategies of the four performs best. Simulation results verify the feasibility of using the compressive sensing as source channel coding on cooperative schemes.