Bayesian inference based resource efficient relay selection scheme for nakagami fading channels

摘要

This paper presents a novel algorithm for relay selection in cooperative cellular networks. Unlike conventional relay selection schemes, this technique restricts the usage of relay nodes for multiple reasons, most significant of which is the wastage of power on the nodes when channel conditions between the two nodes are fair enough and cooperation isn't needed. The said scheme makes the source node intelligent enough to be able to decide whether extra relays are needed in the channel ahead, or if needed then up to what extent it should allow the use of more relays, as to make the communication reliable and at the same time, not waste any resources. Crux of the said decision-making scheme is that ideally, only one relay is utilized during the communication; however, if channel conditions ahead of this relay are not fair enough, the services of other relays are acquired by the relay selection scheme to cooperate and maintain the required Quality of Service (QoS). Analysis of the scheme in various paradigms shows that if the decision-making authority lies with the source node, it gives us a 68% resource efficient relay selection scheme enabling optimal Symbol Error Rate to be achieved within acceptable range of Signal-to-Noise Ratio (SNR).