Performance Analysis of Incremental Relaying in Multi-Hop Relay Networks

Incremental relaying (IR) is shown to improve the end-to-end throughput of multi-hop relay networks. In this paper, two IR strategies under perfect and imperfect channel state information (CSI) scenarios are investigated. In each hop, relays which are able to correctly decode the message from the previous hop form a decoding set. In the first strategy named IR with best-last arbitrary-rest (IR-BLAR), one relay is arbitrarily selected from the decoding set to retransmit the message, except that in the last decoding set the relay which has the highest signal-to-noise ratio (SNR) to the destination is selected for retransmission. In the second strategy called IR with maximum decoding set (IR-MaxDS), in each decoding set one relay which has the largest number of “good” channels in the following hop is selected, except that in the last hop relay selection is the same as IR-BLAR. The performance of these strategies is analyzed in terms of end-to-end outage probability and throughput in both perfect and imperfect CSI cases. Numerical results are presented from analysis which closely match those from simulations. It is shown that by applying IR, end-to-end outage probability and throughput can be significantly improved.