A Local 3D Voronoi-Based Optimization Method for Sensor Network Deployment in Complex Indoor Environments.

Ali Afghantoloee,Mir Abolfazl Mostafavi

doi:10.3390/s21238011

Abstract

Optimal sensor network deployment in built environments for tracking, surveillance, and monitoring of dynamic phenomena is one of the most challenging issues in sensor network design and applications (e.g., people movement). Most of the current methods for sensor network deployment and optimization are empirical and they often result in important coverage gaps in the monitored areas. To overcome these limitations, several optimization methods have been proposed in the recent years. However, most of these methods oversimplify the environment and do not consider the complexity of 3D architectural nature of the built environments specially for indoor applications (e.g., indoor navigation, evacuation, etc.). In this paper, we propose a novel local optimization algorithm based on a 3D Voronoi diagram, which allows a clear definition of the proximity relations between sensors in 3D indoor environments. This proposed structure is integrated with an IndoorGML model to efficiently manage indoor environment components and their relations as well as the sensors in the network. To evaluate the proposed method, we compared our results with the Genetic Algorithm (GA) and the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) algorithms. The results show that the proposed method achieved 98.86% coverage which is comparable to GA and CMA-ES algorithms, while also being about six times more efficient.

Highlights

Advances in sensor technologies increasingly allow efficient and on-the-fly tracking and monitoring of diverse dynamic phenomena for different applications
The existing optimization methods for sensor network deployment and optimization do not consider the complexity of 3D environments and their architectural design
Several strategies have been developed for the local optimization of Innetworks recent years, several strategies been developed the local optimization of sensor based on the Voronoihave structure, which hasfor demonstrated its potential sensor networks onmanagement the Voronoi structure, which has demonstrated its potential esespecially for the based efficient of neighborhood information in sensor networks

Summary

Introduction

Advances in sensor technologies increasingly allow efficient and on-the-fly tracking and monitoring of diverse dynamic phenomena for different applications. For the efficient tracking and monitoring of dynamic phenomena using a sensor network, different issues related to sensor types, sensing models, connectivity, communication, location and coverage as well as efficient assessment of real-time measurements need to be addressed. Among these issues, optimal sensor placement is a prerequisite for efficient monitoring and coverage in a given environment. Other optimization approaches are rarely contextaware and they do not consider the complexity of the environment [2,3,4,5] These methods usually oversimplify sensor models (sensing fields). It is very challenging to consider the presence of diverse obstacles in such environments in the optimization process

Methods

Results

Conclusion