Study On Selection Cooperation Diversity In Wireless Communications

Cooperative communication has received significant interest as an effective way of improving transmission performance. It takes advantage of multiple users'antennas and can be widely used in wireless systems. Selection cooperation is a simple and efficient means in cooperative communication. It uses single relay selection strategy and only one relay which has the best instantaneous channel state will forward the signal to the destination. Compared to other multiple relaying schemes, selection cooperation utilizes all the necessary channel information and achieves full diversity order while greatly reduces the complexity. So it becomes an attractive cooperative method for its flexibility and easy implementation.In this dissertation, we focus on the selection cooperation diversity in wireless communications. Various cooperative schemes based on selection cooperation are proposed and their performances are discussed for different scenarios.Considering the basic scenario with single source and destination, we first develop a distributed cooperative protocol with feedback based on selection cooperation. The best relay is selected in a fully decentralized manner and the protocol can be easily integrated with existing wireless networks. No other information exchanges between relays and destination are needed except acknowledgment frame or negative acknowledgment frame. The diversity-multiplexing tradeoff (DMT) of the proposed protocol as well as common selection cooperation with feedback is derived over Nakagami-m fading channels. The result shows the protocol has significant increase of the spectral efficiency compared with the one without feedback. We also show that the selection cooperation has the same DMT as opportunistic relaying in general fading channels with or without feedback.Next we combine the distributed selection cooperation protocols with automatic repeat request (ARQ) mechanism to develop more powerful cooperative schemes for delay-tolerant wireless networks and analyze their performance from the perspective of diversity-multiplexing-delay (D-M-D) tradeoff. We investigate the general ARQ scheme which directly extends the selection cooperation protocol with single round of feedback to multiple rounds and discuss two cases: memoryless reception and combining reception in the relays. Unlike the opportunistic relaying, we show that using selection cooperation protocol with feedback in ARQ scenario straightforwardly can not achieve the optimal D-M-D tradeoff unless the relays can combine the signals received from previous rounds. Then we propose a simplified scheme which reduces the number of feedbacks to a half and almost achieves the optimal performance. The analysis and simulation results demonstrate that the simplified scheme can improve the reliability of the system by obtaining diversity gain from the ARQ delay and cooperative mechanism.For large networks where direct links are limited in the neighbors of each node by radio coverage and topology, we present a diffusion ARQ protocol based on selection cooperation. The number of potential relays for selection increases as the ARQ process continues and even the nodes that have no direct link with the source could cooperate. The D-M-D tradeoff analysis in the best-worst case demonstrates that the diffusion results in additional diversity benefit. The simulation results show that the diffusion protocol outperforms the general ARQ protocol and achieves better outage probability performance.Finally, we extend the distributed selection cooperation protocol with feedback to multi-source multi-destination cooperative networks and investigate the best relay confliction problem existing in this scenario. In the high signal-to-noise ratio (SNR) regime, due to the fact that sharing the same best relay by multiple source nodes has no influence on the DMT performance for each communication pair, the single relay sharing method is a simple and effective solution. For some practical systems with low or medium SNR, we propose two power allocation algorithms for the shared best relay. One is the max-min method and the other is based on maximizing the successful flow count. The simulation results show that the latter algorithm reduces the complexity greatly and offers better performance.The designs of the cooperative methods in this dissertation take into account both the analytic performances and the practical issues. So the protocols are easy to carry out in practice with the characteristics of simplicity. The proposed schemes and their performance derivations contribute to the development and applications of cooperative diversity in wireless networks.