Improved range-free localization algorithm based on reliable node optimization and enhanced sand cat optimization algorithm

• Published in: The Journal of supercomputing Vol. 79; no. 18; pp. 20289 - 20323
• Main Authors: Sun, Haibin, Tian, Meng
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.12.2023; Springer Nature B.V
• Subjects: Accuracy; Algorithms; Compilers; Computer Science; Interpreters; Localization; Localization method; Network topologies; Nodes; Optimization; Processor Architectures; Programming Languages; Sand; Wireless sensor networks; Wireless sensor network; RESCO algorithm; Reliable node; Localization; DV-Hop
• ISSN: 0920-8542; 1573-0484
• DOI: 10.1007/s11227-023-05436-4

For wireless sensor networks, it is crucial to determine the geographic area where events occur, and the localization of unknown nodes has become one of the challenging problems today. The Distance Vector-Hop (DV-Hop) algorithm has gained attention as a popular localization method for wireless sensor networks. However, it faces limitations in irregular areas and regions with unevenly distributed nodes, as the accuracy of hop distance calculation is heavily impacted by network topology. For irregular networks under the influence of obstacles and voids, the proposed algorithm optimizes the hop spacing of reliable beacon nodes within a limited number of hops and designs an enhanced sand cat optimization (ESCO) algorithm to solve the localization problem. In the simulation tests, benchmark functions with different characteristics are used to verify the convergence speed and accuracy of the ESCO algorithm. In addition, the proposed algorithm based on reliable node optimization and enhanced sand cat optimization (RESCO DV-Hop) is also applied to the node localization process and compared with five recent algorithms. The experimental results show that the proposed algorithm achieves the best localization accuracy under C-, H-, and X-type network structures. More precisely, the localization accuracy is improved by around 58, 53, and 43%, respectively, compared with the original algorithm, in the best case.