Enhancing Network Reliability: Exploring Effective Strategies for Coverage-Hole Analysis and Patching in Wireless Sensor Networks

• Published in: Wireless personal communications Vol. 134; no. 1; pp. 487 - 517
• Main Authors: Chowdhuri, Rajib, Barma, Mrinal Kanti Deb
• Format: Journal Article
• Language: English
• Published: New York Springer US 2024; Springer Nature B.V
• Subjects: Algorithms; Boundaries; Communications Engineering; Computer Communication Networks; Effectiveness; Efficiency; Engineering; Graph neural networks; Machine learning; Network reliability; Networks; Nodes; Optimization; Patching; Performance evaluation; Resource utilization; Sensors; Signal,Image and Speech Processing; Wireless sensor networks; Coverage-hole; Patching; WSN; Patch position; Game optimization; Graph neural network
• ISSN: 0929-6212; 1572-834X
• DOI: 10.1007/s11277-024-10933-7

In the context of wireless sensor networks (WSNs), the study focuses on addressing coverage holes, representing areas where sensor nodes lack sufficient coverage due to factors such as node failure, energy depletion, or environmental obstacles. The research proposes an innovative approach utilizing advanced deep learning techniques to analyze and mitigate coverage holes in WSNs. The Improved Social Group Optimization (ISGO) algorithm quantifies coverage hole areas within the Region of Interest (RoI), offering insights for targeted patching. Subsequently, the Decentralized Game Optimization (DGO) algorithm is employed to precisely identify coverage hole boundaries, optimizing resource utilization for effective patching strategies. This research aims to enhance the reliability and comprehensiveness of monitoring in WSNs by addressing the challenges posed by coverage holes. Based on the identified coverage hole and its boundaries, we devise a comprehensive patching strategy to address the coverage deficiency. By strategically placing new sensor nodes within the RoI, we aim to fill the identified patch positions and ensure continuous coverage. The selection of these patch positions is guided by the revisiting graph neural network (RGNN), taking into consideration factors such as connectivity, energy efficiency, and overall network performance. To validate the effectiveness of proposed approach, we conduct extensive performance evaluations under various scenarios. By addressing coverage holes and deploying optimal patching strategies, our approach significantly improves the overall performance and efficiency of WSNs. The ISGO + DGO + RGNN technique presents substantial enhancements in coverage-hole detection and patching. Notably, it achieves significant reductions in hole detection time (87–99.6%), requires fewer patching sensors (56–70%), exhibits lower time complexity (26–83%), and shows decreased energy overhead (34–82%) compared to existing techniques. Moreover, the proposed technique consistently attains higher coverage rates (23–29%), affirming its superior efficiency and effectiveness in addressing the limitations of current methods for WSNs.