Power-Aware Processors for Wireless Sensor Networks

Abstract

Today, wireless sensor networks (WSNs) enable us to run a new range of applications from habitat monitoring, to military and medical applications. A typical WSN node is composed of several sensors, a radio communication interface, a microprocessor, and a limited power supply. In many WSN applications, such as forest fire monitoring or intruder detection, user intervention and battery replenishment is not possible. Since the battery lifetime is directly related to the amount of processing and communication involved in these nodes, optimal resource usage becomes a major issue. A typical WSN application may sense and process very close or constant data values for long durations, when the environmental conditions are stable. This is a common behavior that can be exploited to reduce the power consumption of WSN nodes. This study combines two orthogonal techniques to reduce the energy dissipation of the processor component of the sensor nodes. First, we briefly discuss silent-store filtering MoteCache. Second, we utilize Content-Aware Data MAnagement (CADMA) on top of MoteCache architecture to achieve further energy savings and performance improvements. The complexity increase introduced by CADMA is also compensated by further complexity reduction in MoteCache. Our optimal configuration reduces the total node energy, and hence increases the node lifetime, by 19.4% on the average across a wide variety of simulated sensor benchmarks. Our complexity-aware configuration with a minimum MoteCache size achieves not only energy savings up to 16.2% but also performance improvements up to 4.3%, on the average.