Toward a Realistic Approach for the Deployment of 3D Wireless Sensor Networks

Deploying wireless sensor networks (WSNs) in real-world applications is still a challenging problem. Indeed, the number, locations, and orientations of directional sensors determine the topology of the WSN, which will further influence its overall cost and performance. Existing works have assumed simple sensor behavior and/or ideal environmental factors; thus, often they do not produce practical results. On the contrary, in this paper, we revisit the 3D WSNs deployment problem while considering realistic assumptions regarding the modeling of both the sensors and the environment. More precisely, we propose a Bresenham line-of-sight based realistic coverage model for 3D environments. This latter is used to re-formalize the 3D WSNs deployment problem while considering a realistic spatial model of the environment. The problem is then solved using a multi-objective genetic algorithm endowed with new adaptive and guided genetic operators. Moreover, we enhance the performance of the proposed approach by introducing two optimization techniques, namely: search space reduction and sampling-based evaluation. We show the effectiveness and efficiency of the proposed approach through extensive simulations.