Robust extreme energy efficient sensor networks

Despite much research on sensor networks in general and on event-response applications in particular, simultaneously achieving the desired timeliness of response, longevity, and wide-area coverage still remains a significant challenge. In this paper, we present a robust, flexible, and extremely energy efficient sensor networking system solution with the goal of enabling orders of magnitude improvements in sensor network lifetime for time-stringent event-response applications with wide area coverage requirement such as maritime domain awareness and environmental monitoring. We first present a cross-layer hierarchical energy adaption framework, a main departure from traditional approach of single-component adaptation and efficiency paradigm. The proposed solution consists of the following components: duty-cycle optimized radio, low-power adaptive multimodal MAC (LPA-MAC), and cross-layer adaptive reconfiguration (CARE). This paper focuses on the cross-layer design aspects of LPA-MAC and CARE. For LPAMAC, we conduct a thorough investigation on the relationship among duty cycle, synchronization mode and traffic load. Contrary to common belief, we find that there exists a range of energy-optimal duty cycles for a given traffic load. The resulting relationship provides us with critical guidelines on how to adaptively select near optimal operational mode so that only the least possible energy is consumed for given traffic loads. For CARE, a concept of adaptively reconfiguring a backbone structure is introduced and evaluated. Preliminary simulation results indicate more than two orders of magnitude in energy saving over an ldquoalways-onrdquo system with comparable timeliness.