A Hybrid Voronoi Tessellation/Genetic Algorithm Approach for the Deployment of Drone-Based Nodes of a Self-Organizing Wireless Sensor Network (WSN) in Unknown and GPS Denied Environments

Abstract

Using autonomously operating mobile sensor nodes to form adaptive wireless sensor networks has great potential for monitoring applications in the real world. Especially in, e.g., disaster response scenarios—that is, when the environment is potentially unsafe and unknown—drones can offer fast access and provide crucial intelligence to rescue forces due the fact that they—unlike humans—are expendable and can operate in 3D space, often allowing them to ignore rubble and blocked passages. Among the practical issues faced are the optimizing of device–device communication, the deployment process and the limited power supply for the devices and the hardware they carry. To address these challenges a host of literature is available, proposing, e.g., the use of nature-inspired approaches. In this field, our own work (bio-inspired self-organizing network, BISON, which uses Voronoi tessellations) achieved promising results. In our previous approach the wireless sensors network (WSN) nodes were using knowledge about their coverage areas center of gravity, something which a drone would not automatically know. To address this, we augment BISON with a genetic algorithm (GA), which has the benefit of further improving network deployment time and overall coverage. Our evaluations show, unsurprisingly, an increase in energy cost. Two variations of our proposed GA-BISON deployment strategies are presented and compared, along with the impact of the GA. Counter-intuitively, performance and robustness increase in the presence of noise.

Keywords

• Voronoi centroids
• genetic algorithm
• particle swarm optimization
• drones
• drone swarms
• swarming