Optimal Resource Allocation and Performance Comparison of Half-Duplex Relay Strategies

Although the established capacity results for the classic three-node relay channel rely on the assumption of fullduplex relay nodes, it's practically difficult for radio transceivers to support simultaneous signal transmission and reception. Timedivision half-duplex relay strategies have thus been proposed to address this limitation. Amplify-and-forward (AF) and decode-and-forward (DF) are two well-known half-duplex relay strategies in the literature. Later, DF approach was extended by the so-called variable-rate two-phase relay strategy, in which the transmission rates for both phases are not necessarily the same. More recently, a novel half-duplex relay strategy based on the information theoretical concepts of broadcast channel (BC) and multiple access channel (MAC) models is proposed. In this approach, the source broadcasts distinct data flows to the destination and the relay respectively during the first phase, and in the second phase the relay forwards the received information whereas the source sends another data flow to the destination. In this paper, we will first develop optimal resource allocation, in terms of power and time, for these half- duplex relay strategies, and then compare their performance under various link conditions.