M-SPOT: A Hybrid Multiobjective Evolutionary Algorithm for Node Placement in Wireless Sensor Networks
Alfredo J. Pérez · 2018
We address the problem of the placement of static sensors and relays to monitor specific locations in an area assuming a single-tiered wireless sensor network model with limited communication and sensing constraints. We present a multiobjective optimization model with two conflicting objectives: total number of devices used in the placement and total energy dissipated by the placement. To optimize the model, we propose the Multiobjective Sensor Placement Optimizer (M-SPOT) algorithm, which is a hybrid evolutionary algorithm that combines the Non-Sorting Genetic Algorithm 2 (NSGA2) algorithm with local search heuristics. We evaluate the performance of M-SPOT by simulating the placement of sensors and relays. We found that the utilization of local search heuristics greatly contribute to find better placements when compared to the NSGA2 algorithm.