Algorithms for Detecting and Refining the Area of Intangible Continuous Objects for Mobile Wireless Sensor Networks

Detecting the intangible continuous object (ICO) is a significant task, especially when the ICO is harmful as a toxic gas. Many studies used steady sensors to sketch the contour and find the area of the ICO. Applying the mobile sensors can further improve the precision of the detected ICO by efficiently adjusting the positions of a subset of the deployed sensors. This paper proposed two methods to figure out the area of the ICO, named Delaunay triangulation with moving sensors (MDT) and convex hull with moving sensors (MCH). First, the proposed methods divide the sensors into ICO-covered and ICO-uncovered sensors. Next, the convex hull algorithm and the Delaunay triangulation geometric architecture are applied to figure out the rough boundary of the ICO. Then, the area of the ICO is further refined by the proposed sensor moving algorithm. Simulation results show that the figured out area sizes of MDT and MCH are 135% and 102% of the actual ICO. The results are better than the planarization algorithms Gabriel Graph (GG) and Delaunay triangulation without moving sensors, that amount to 137% and 145% of the actual ICO. The simulation also evaluates the impact of the sensors' moving step size to find the compromise between the accuracy of the area and the convergence time of area refinement.

Automated summary

This paper proposes two methods for detecting the area of ICO and improves the detection accuracy by utilizing mobile sensors. The simulation results demonstrate that the proposed methods yield more accurate area sizes compared to other algorithms.