The implementation of cooperative diversity with relays has advantages over point-to-point multiple-input multiple-output (MIMO) systems, in particular, overcoming correlated paths due to small inter-element spacing. A simple transmitter with one antenna may exploit cooperative diversity or space time coding gain through distributed relays. In this paper, similar distributed transmission is considered with the golden code, and the authors propose a new strategy for relay selection, called the maximum-mean selection policy, for distributed transmission with the full maximum-likelihood (ML) decoding and sphere decoding (SD) based on a wireless relay network. This strategy performs a channel strength tradeoff at every relay node to select the best two relays for transmission. It improves on the established one-sided selection strategy of maximum-minimum policy. Simulation results comparing the bit error rate (BER) based on different detectors and a scheme without relay selection, with the maximum-minimum and maximum-mean selection schemes confirm the performance advantage of relay selection. The proposed strategy yields the best performance of the three methods.
Predictive Relay Selection: A Cooperative Diversity Scheme Using Deep Learning
