Cooperative Incremental Redundancy Hybrid Automatic Repeat Request Strategies for Multi-Source Multi-Relay Wireless Networks

In this letter, we present several cooperative incremental redundancy hybrid automatic repeat request strategies for the slow-fading half-duplex time slotted multiple access multiple relay channel. At a given retransmission round, the proposed strategies main steps are: 1) the destination acknowledges the successful decoding of the sources via limited feedback control channels; 2) the relays inform the destination about the result of their own decoding via forward coordination control channels; 3) the destination schedules the re-transmission in order to minimize the common outage probability of the system with the overall goal to maximize the long-term aggregate throughput; and 4) the scheduled relay cooperates with a subset of the successfully decoded sources (implicitly or explicitly indicated by the destination) applying either distributed coding or joint network channel coding. The analytical derivation of the individual (per source) and common outage probability plays an essential role in the scheduling algorithm and the performance evaluation. Numerical results show that the proposed strategies with control channels significantly outperform the one without control channels even when the impact of the control overhead is taken into account.

I. Introduction

The Multiple Access Multiple Relay Channel (MAMRC), denoted by -MAMRC, is a model for network topologies where independent sources (or users) communicate with a single destination in the presence of dedicated relays. This channel model could be seen as a generalization of the relay channel and the multiple access relay channel. In this letter, we consider a restrictive version of MAMRC, namely slow-fading half-duplex time-slotted Orthogonal MAMRC (OMAMRC) where the multiple access is orthogonal (TDMA), transmission is scheduled in consecutive slots, links are subject to slow fading, relays operate in half-duplex mode, and apply a Selective Decode-and-Forward (SDF) cooperative strategy. An SDF relay tries to decode the sources and forwards a function of the successfully decoded ones (based on Cyclic Redundancy Check (CRC) codes inserted in the sources’ packets). SDF has several advantages: (1) It prevents decoding error propagation from the relays to the destination; (2) It reduces the energy consumption at the relays and the level of interference in the network; (3) It provides a power gain compared to Non-Selective Decode-and-Forward (NSDF), where the relay cooperates only if it successfully decodes all the sources. In theory, there is no difficulty for a receiving node to know which sources are involved in the transmitted relay signal, since it can review all possible combinations and choose the one leading to the highest likelihood (under maximum likelihood decoding). In practice, SDF often relies on specific SDF forward coordination control channels (additional coordination bits embedded in relays’ transmission) to reduce the decoding complexity of the receiving nodes.