Dynamic Power Management Design for Energy Harvesting Wireless Sensor Networks

Abstract

A Wireless Sensor Network (WSN) is a network of spatially distributed, battery- newlinepowered sensor nodes that monitor physical or environmental conditions, such as newline newlinetemperature, humidity, pressure, light, or motion. These sensor nodes communicate newlinewirelessly to relay data to a central processing system or base station for analysis. newlineWSNs are commonly used in a variety of applications, such as environmental newlinemonitoring, industrial automation, health care, military surveillance, smart cities, newlineand agriculture. Wireless sensor networks facilitate the monitoring and controlling newlineof physical environments from remote locations, with enhanced accuracy. They have newlineapplications in a variety of fields such as environmental monitoring, military newlinepurposes and gathering of sensing information in inhospitable locations. So typical newlineapplications include, but are not limited to, data collection, monitoring, surveillance newlineand medical telemetry. They are dense networks for environment sensing and data newlinecollection. Sensors are equipped with both data processing and communication newlinecapabilities. They measure different parameters from the environment and transform newlinethem to electric signals. The prime advantage of sensors is their capability to operate newlineunattended in harsh environments. Wireless sensor networks have a high utility in a newlinevariety of industrial, medical, consumer and military applications. They can be newlineclassified into three main classes based on the methods of acquiring and propagating newlinesensor data. The first class comprises of processes requiring constant monitoring, newlinewhere sensor data is acquired from a number of remote points and forwarded to a newlinedata collection center on a periodic basis. The second class is event driven, in which newlineone or more crucial variables are to be monitored but transmitted only when a newlinecertain threshold is reached. newlineWireless sensor networks have gained rapid worldwide attention in recent years, newlinemainly with the proliferation in Micro-Electro-Mechanical Systems (MEMS) newlinetechnology which has facilitated the development of small and low cost sensors. But newlinethe sensor node lifetime exhibits a strong dependence on its battery life. newline newline(iv) newline newlineIn most cases, sensor nodes are equipped with limited power sources and newlinereplenishment of power may be practically impossible since they are deployed in newlineharsh environments. Power management and power conservation are critical newlinefunctions for sensor networks and the need to design power aware protocols and newlinealgorithms is quite significant. No standard protocols and algorithms exist currently newlineand an integration of many techniques, considered to tackle this problem, is highly newlineessential. It was also found from previous literatures that factors like Delay, newlineThroughput, Stability and Delivery ratio affect the performance of a wireless sensor newlinenetwork apart from Energy efficiency. Hence any new scheme needs to be evaluated newlinebased on these factors. newlineThe general framework for energy efficient data acquisition is based on a duty cycle newlineapproach requiring the sensor to be switched OFF during idle time. Dynamic Power newlineManagement is an effective tool in reducing the system power consumption without newlinedegrading the performance. The quality of wireless channels or the network newlinetopology can change rapidly even in a static network. Routing protocols should newlinequickly detect these dynamics and take measures to maintain robust and efficient newlinerouting paths. This needs to be considered in the design of a wireless sensor newlinenetwork. Moreover a cross layer design on power aware communication, newlineconsidering the time varying nature of wireless channels, is needed to optimize the newlineenergy usage. In this work, all the above mentioned ideas are incorporated. Here newlineeach sensor node judiciously accesses the wireless medium and communication newlineactivity is reduced for those sensor nodes, under poor channel conditions. Hence the newlinetime varying nature of wireless channels are taken into account in this work, in the newlineoptimization of energy management in a wireless sensor network. newline