Improved Partner Selection Method For Cooperative Network

In cooperative communication, partner selection is an important step in realizing cooperative protocol and a key factor which affect the performance of the system. In this paper, the existing partner selection algorithms were summarized. Based on the existing algorithms, two new improved partner selection algorithms were proposed for two real scenarios of cooperative wireless network.First, for cooperative wireless sensor network, relay node competition problem under this scenario was analyzed and a distributed relay multiplexing partner selection method was promoted for solving this problem. A system model which allows cooperative links to multiplex the relay node was given. Maximum cooperation gain was obtained by optimizing relay node's power allocation with Lagrange Method. The condition and optimum power allocation method when system can get BER gain or power gain was analyzed. Theory analysis and simulation results prove that the new method can satisfy the BER requirement of links better and reduce total power consumption.Then, effect was spent on a cooperative cellular network where IDMA superposition modulation was applied. It is convinced that superposition modulation can improve time slot efficiency. However, the application of superposition modulation has brought new problems for partner selection in a cooperative cellular network. In this paper, first the expression of outage probability for a cooperative network based on IDMA superposition modulation was derived. Then the condition when IDMA superposition modulation could outperform direct transmission was analyzed. By changing existing outage gain matrix into a symmetric new one and applying Hungary and WLF algorithm, a centralized optimal and a distributed sub-optimal partner selection strategies were proposed for superposition modulation CMA cellular network base on the modified gain matrix. Simulation results shows that the proposed partner selection strategies could reduce the average outage of the network effectively.