OTGsim: Simulation of an Off-the-Grid Radar Network with High Sensing Energy Cost

Many sensor network studies assume that the energy cost for sensing is negligible compared with the cost of communications or computing. Opportunities exist to deploy sensor networks utilizing active sensors with a high energy cost such as radar. For a node utilizing radar as its primary sensor, the actual sensing procedure is the main power consumer. In the worst case almost 50% of the power is consumed by the sensing procedure, while only 3% is used for communication, the remainder is consumed by the computing platform. In this paper we examine a wireless sensor network composed of short range radars used to monitor rainfall. These short-range radar nodes are designed to be deployed as part of an ad-hoc network and to limit their reliance on existing infrastructure. We refer to these networks as "off-the-grid" (OTG) weather radar networks. Independence of the wired infrastructure (power or communications) allows OTG networks to be deployed in specific regions where sensing needs are greatest, such as mountain valleys prone to flash-flooding, geographic regions where the infrastructure is susceptible to failure, and underdeveloped regions lacking urban infrastructure. We present a simulation based investigation of such an OTG sensor network. We focus on power management and energy harvesting for the network. We use these simulations to demonstrate how geographic location, battery capacity, optimization of power consumption, and node density have an impact on the performance and operational lifetime of such a sensor network. In addition to these simulations, we present the design and implementation of an OTG prototype sensor node. Experiences and data gained from the operation of this node are used as input parameters for the simulations.