Link adaptation for energy constrained networks

Wireless sensor networks (WSNs) are a nascent technology that has recently garnered a significant amount of interest in academia and industry. Nodes in a sensor network often only have a finite amount of energy that can be used for sensing, signal processing, and communication. Understanding the trade off between battery life, signal processing and wireless communication is of paramount importance if we hope to maximize the operational lifetime of the network.;Relay terminals are often used in tandem in sensor networks to lessen nodal communication burden. In this light we investigate the problem of power allocation amongst nodes in a relay network in order to maximize the overall achievable rate using link adaptive transmission protocols. We focus on the physical layer characteristics and implementation issues of link adaptation in order to develop a bit-level simulator needed to accurately model the rate performance of such a system.;Optimal power allocation values, power adaptation policies, and switching levels for several link adaptive policies over a broad class of Rician fading channels are calculated. Furthermore, the maximum achievable rate for two and three link relay networks using our bit-level simulator and optimal power allocation values for collocated channel distributions is simulated. An overall achievable rate comparison between several link adaptive protocols is also investigated.;Finally, we show that discrete-rate continuous-power adaptation is a superior candidate to other link adaptive protocols and that an appropriate design facilitates its incorporation as a communication protocol for energy constrained relay networks.