Rateless Coding Based Cooperative Transmission For Cognitive Relay Systems

The performance of future wireless networks is restrained by the limited spectrumresource and the unreliable wireless transmission. With the mutual penetration of co-operative relaying and cognitive radio, a combination of them will provide a promisingperspective to improve spectrum utilization, transmission reliability, and network con-nectivity in future wireless networks. With complicated spectrum environment and time-varying network state, a key issue is how to provide reliable transmission using errorcontrol techniques and guarantee the quality of service in cognitive relay systems. In pre-vious work, error control techniques for cooperative transmission and cognitive radio arestudied separately. Moreover, less attention has been paid to the interaction between co-operative relaying and cognitive radio. This dissertation investigates the way of improv-ing the spectrum efciency of cooperative transmission, spectrum access, and cognitiverelaying with the aid of rateless codes.Using probability theory, Markov decision process, and order statistics, this disserta-tion address the folloing four issues. Firstly, we establish a unified analytical frameworkbased on two novel metrics, namely forward redundancy and feedback redundancy, andpresent a comparative study on Automatic Repeat reQuest (ARQ) and rateless codingfrom the point of view of redundancy. In particular, we conduct a fair comparison of bothschemes over point-to-point and broadcast fading channels, respectively. The comparisonindicates that rateless coding is capable of beating ARQ completely at low signal-to-noiseratios in broadcast communications. Secondly, by jointly combining rateless coding andsuperposition coding, we propose a new cooperative relay transmission scheme for two-hop relay systems, which can also be applied to multihop relay networks. With optimalpower allocation, the receiver is able to strike a balance between energy accumulationand information accumulation. The proposed scheme can achieve about20%and40%throughput gains in two-hop and multihop relay systems, respectively. Thirdly, we pro-pose a rateless-coding-based dynamic spectrum access scheme for cognitive radio. Inparticular, we design the optimal spectrum sensing policy based on the theory of PartiallyObservable Markov Decision Process (POMDP). Moreover, we employ rateless code toaggregate discontinuous spectrum holes in diferent dimensions. Finally, we propose arateless coded cooperative transmission scheme for cognitive relay systems. Particularly, the transmission reliability can be increased using rateless codes and less interference canbe induced by using relay transmission. In addition, with cognitive capability, the relaynodes can explore and access diferent spectrum holes, thereby improving the networkconnectivity. We analyze the average end-to-end throughput and study the impact of pri-mary users’ behaviors and wireless channel parameters on the system performance so asto determine the optimal system parameter.This dissertation provides an in-depth analysis of the property of rateless codes andapplies rateless codes to cooperative relaying and cognitive radio. We employ ratelesscodes to provide reliable transmission with complicated spectrum environment and time-varying network state. The schemes proposed in this dissertation can achieve substantialperformance gain. Moreover, this dissertation provides a theoretical foundation and prac-tical techniques for future wireless networks.