PSLDV-Hop: a robust localization algorithm for WSN using PSO and refinement process

• Published in: PeerJ. Computer science Vol. 11; p. e2770
• Main Authors: Kaur, Bhupinder, Prashar, Deepak, Khan, Arfat Ahmad, Kadry, Seifedine, Kim, Jungeun
• Format: Journal Article
• Language: English
• Published: United States PeerJ. Ltd 18.07.2025; PeerJ Inc
• Subjects: Algorithms; Localization; Localization Error; Mathematical optimization; Optimization; Sensor Nodes; Sensors; Wireless sensor networks; Wireless telecommunications equipment; WSN; United States; AI; Sensor Nodes; WSN; Localization; Localization Error; Optimization
• ISSN: 2376-5992; 2376-5992
• DOI: 10.7717/peerj-cs.2770

In various areas, wireless sensor networks (WSNs) are popular for achieving goals related to security in buildings when there is fire, in military areas to know the position of terrorists in moles and to observe the behavior of animals in forest areas. All these objectives can be achieved only when the position of the sensor is known to the base station, which helps to achieve the appropriate action in unwanted situations. The controlling point is the base station, which would be able to take action only in case the correct position of the unwanted event is known to the base station. Researches have designed various localization/positioning approaches but still have some challenges related to the accuracy of sensor nodes in localization. Distance vector hop is a popular localization algorithm. Its dependence on the estimated average size of a hop results in a significant localization error. This work suggests an improved algorithm combining a refinement procedure with particle swarm optimization, called DVHOP-PSO. This improved algorithm, called PSLDV-Hop, uses exact anchor sensor node coordinates and fractional hop count information to correct estimated distances. By utilizing an improved iterative evolution algorithm, the PSLDV-Hop algorithm reduces localization errors by achieving a higher degree of accuracy in node localization. Simulation results demonstrate their superiority over other classical improved algorithms and the original distance vector hop. The simulation of this approach is done using the MATLAB tool by considering different parameters such as the number of anchor nodes, number of sensor nodes, area, and range of sensor nodes. Integrating particle swarm optimization with distance vector hop, the proposed localization algorithm consistently outperforms conventional methods, showcasing significant percentage improvements . The suggested algorithm consistently performs better than all other approaches at ranges 20 and 40. Overall, the suggested method performs noticeably better than distance vector hop at range 40, especially when range grows by up to 65%. Additionally, across communication ranges of 20, 30, and 40 units, the proposed algorithm consistently outshines PSO-DV-Hop and GA-DV-Hop, exhibiting notable percentage improvements in localization accuracy.