Efficient meta-heuristic approaches in solving minimal exposure path problem for heterogeneous wireless multimedia sensor networks in internet of things

• Published in: Applied intelligence (Dordrecht, Netherlands) Vol. 50; no. 6; pp. 1889 - 1907
• Main Authors: Binh, Nguyen Thi My, Binh, Huynh Thi Thanh, Van Linh, Nguyen, Yu, Shui
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.06.2020; Springer Nature B.V
• Subjects: Artificial Intelligence; Computer Science; Evolutionary algorithms; Genetic algorithms; Heuristic methods; Internet of Things; Linearity; Machines; Manufacturing; Mechanical Engineering; Multimedia; Particle swarm optimization; Processes; Wireless networks; Evolution algorithm; Internet of things; Heterogeneous wireless multimedia sensor network; Minimal exposure path; Directional sensing coverage model; Particle swarm optimization algorithm
• ISSN: 0924-669X; 1573-7497
• DOI: 10.1007/s10489-019-01628-9

One of the well-known methods for evaluating Heterogeneous wireless multimedia sensor networks (HWMSNs) in Internet of Things have drawn attention of the research community because this type of networks possesses great advantages of both coverage and performance. One of the most fundamental issues in HWMSNs is the barrier coverage problem which evaluates the surveillance capability of the network systems, especially those designed for security purposes. Among multiple approaches to solve this issue, finding the minimal exposure path (MEP), which corresponds to the worst-case coverage of the network is the most popular and efficient way. However, the MEP problem in HWMSNs (hereinafter heterogeneous multimedia MEP or HM-MEP) is specifically complex and challenging with the unique features of the HWMSNs. Thus, the problem is then converted into numerical functional extreme with high dimension, non-differential and non-linearity. Adapting to these features, two efficient meta-heuristic algorithms, Hybrid Evolutionary Algorithm (HEA) and Gravitation Particle Swarm Optimization (GPSO) are proposed for solving the problem. The HEA is a hybrid evolutionary algorithm in combination with local search while the GPSO is a novel particle swarm optimization based on the gravity force theory. Experimental results on extensive instances indicate that the proposed algorithms are suitable for the HM-MEP problem and perform well in term of both solution accuracy and computation time compared to existing approaches.