Relaying, MultiHop, and Cooperative Communications

This chapter discusses basics of relaying, multi-hop transmission, and cooperative communications. We start with one source, one destination, and one relay. Various processing schemes (xF) at the relay (amplify-and-forward AF, decode-and-forward DF, compress-and-forward CF) together with various destination signal combining schemes, result in multi-hop xF, split-combine xF, diversity xF, nonorthogonal diversity xF, and intersymbol interference xF. Full-duplex and half-duplex constraints are considered. In two-phase protocols with parallel relays, cooperation schemes of the relays include relay selection, distributed beamforming, distributed space-time coding, coded cooperation, and mutual information accumulation using Fountain codes. Multihop relaying in larger networks require joint routing and resource allocation. Dijkstra and Bellman-Ford algorithms provide shortest-path routing. More generally we distinguish between proactive and reactive protocols. Noteworthy protocols include source routing, link-state routing (e.g., Optimized Link State Routing protocol OLSR), distance-vector routing (e.g., Destination-Sequenced Distance Vector DSDV and Ad hoc On-demand Distance Vector AODV), geography-based routing, and hierarchical routing. Node mobility is exploited in epidemic routing. Data-driven routing, e.g., directed diffusion, is used for sensor data. Stochastic network optimization (backpressure algorithms) forwards information without establishing explicit routes. Routing changes when nodes collaborate during each hop for better diversity and energy efficiency, e.g., in anypath routing (exploiting the broadcast effect), edge-disjoint shortest-path routing, and routing with nodes that use energy accumulation or mutual information accumulation. We finally discuss two-way relaying and general network coding, where nodes form combinations of received codewords, according to an encoding vector, or use over-the-air combination of signals (compute-and-forward).