Cross-layer protocols for energy-efficient wireless sensor networking

The Army's next generation distributed sensor networks provide stand-off situational awareness for future troops. A major factor for the performance and lifecycle of these sensor networks is their ability to conserve battery power. As advances in RF components reach the physical limits of energy efficiency, new network protocols operating at the link layer and above hold the greatest opportunities for additional improvements in energy efficiency. Recently, a number of power-saving solutions have been proposed that separately consider power-consumption of media access control (MAC) scheduling and routing algorithms, but have not considered the potential benefit of cross-layer optimization across these algorithms. In this paper, we present an approach that integrates pseudo-dynamic scheduling at the link-layer with diversity routing at the network-layer. The link-layer protocol also provides detailed control of the physical-layer's radio state. Our work focuses on reducing the energy loss due to idle listening, control signaling, congestion hot-spots, and packet collisions. We present analytical and simulation results that demonstrate the increased energy efficiency of this class of cross-layer protocols examine the throughput and latency impacts and define how parameters should be set to optimize the end-to-end performance. We anticipate that the technology presented here will have broad application to army sensor networks as well as public commercial wireless systems and plant automation.