Compress-and-Forward System Model And Optimization For Multi-relay Networks Based On Gaussian Sources

With the rapid development of wireless communication technology,data transmission rate,network coverage and information quality need to be improved.Wireless relay technology has been widely used in all kinds of wireless communication systems because it can effectively combat fading in wireless channel,increase transmission reliability and improve the system performance.At present,the research of relay technology mainly focuses on three aspects: forwarding strategy,power allocation and relay selection.The relays still often use simple amplify-and-forward strategy when the source is analog signal.In this paper,we present a new compress-and-forward system model and optimization method for Gaussian source.The system model can describe such kind of relay networks with such characteristics in practice: the source is limited by environment or processing ability and can only send analog signal,while the relays can carry out complex distributed source coding and channel coding.The main work of this paper is as follows:First,the compress-and-forward system model with single-source,multi-relay isomorphism is studied,assuming that the signal-to-noise ratio(SNR)of each relay is same.We propose a new compress-and-forward system model,in which the analog Gaussian signal is transmitted by the source,and then the distributed compression coding of the analog signal is carried out at the relays.We use the CEO theory of distributed source coding to establish the rate-distortion function,and combine the Shannon channel capacity theory to establish the optimal model of the system.Under the condition of total power limitation,the power allocation between the source and relays is carried out to get the maximum SNR performance at the destination.Finally we compare the SNR performance of the proposed CF system with that of the AF system under the same total power condition,the simulation shows that the proposed CF system can get better SNR performance.Then,we further study the CF system with isomerism form relays,which means the SNR of each relay is different.The rate-distortion function of each relay after distributed source coding is derived,then the optimal model is proposed.Finally we proposed an iterative algorithm to carry on relay selection and power allocation between the source and relays to get the maximum SNR performance at the destination.Finally,the CF system model of two correlated sources is studied.We derive the rate-distortion function of two-dimensional vector correlation source after distributed source coding at relays and also combine with the Shannon channel capacity theory to establish the optimization model to maximize the SNR performance at the destination.In addition,the influence of the correlation between the sources is further studied.The simulation results show that the destination can get better SNR performance under the same total power condition when considering the correlation between the sources.