Proceedings of the 82nd Military Operations Research Society (MORS) Symposium, Operational Energy working group, Alexandria, VA (June 2014)

Wireless ground sensor networks provide superior situational awareness in operational areas such as around forward operating bases (FOB) and command outposts (COP), but their deployment is complicated by unpredictable signal traffic patterns, limited battery life, and physical obstacles that interfere with signal propagation. Sub-optimal sensor network layouts create transmission bottlenecks and force nodes to drain batteries faster than necessary. Currently, personnel must rely on experience and trial and error to achieve an optimal configuration, which may not be practical in hostile territory or during time-sensitive missions. We have designed a prototype tactical decision aid (TDA) for Sensor Placement Reasoning using an Evolutionary Algorithm and Digital elevation maps (SPREAD) with a robust, intelligent approach to calculate optimal sensor layouts. Our solution consists of an evolutionary algorithm to find the optimal sensor placement for a given 3D terrain based on tradeoffs between coverage, number of sensors, and connectivity. We use a heuristic approach to model and predict possible run-time behavior of the mesh networks in the field. We use a two-stage hybrid model to improve our signal propagation path loss calculations. Depending on the resolution of the terrain elevation map and the geographic scale of the network, we select either the Terrain Integrated Rough Earth Model (TIREM) or a custom hybrid of the free space and plane Earth models to compute the path loss between nodes. These optimize layouts to (1) maximize the operational lifetime based on batteries’ energy levels and usage patterns; (2) minimize the number of relay radios required to provide the needed connectivity; (3) minimize latency (maximizing speed) from all sensors to the gateway node; and (4) combine multiple criteria to improve reliability in the event of relay failure. We have tested this TDA in simulation using various DTED level 2 terrain maps with promising results.