Optimize the Network Topology in Underwater Sensor Networks (UWSNs) to Improve the Localization

Abstract

Underwater Wireless Sensor Networks (UWSNs) offer significant advantages due to their wide-ranging applications, largely dependent on the placement of sensor nodes. Effective algorithms for locating or identifying underwater target objects via UWSNs hinge on the sensor nodes' ability to accurately determine their surroundings, making this a key area of research. Enhancing localization for large-scale mobility in UWSNs is challenging due to adverse aquatic conditions, considerable node mobility, and the extensive scale of the network. Many experts have refined localization algorithms or developed new methods to improve target node accuracy, advancing the field. This work improves the node localization in UWSNs. It uses Time Difference of Arrival (TDOA) to measure distances and the Red Vulture Optimization Algorithm (RVOA) for precise localization. The method also employs Euclidean distances and window prediction to reduce errors and delays. The node mobility model predicts velocity and position over time. The proposed work is compared with existing methods like MPL, GA-SLMP, SLMP, and LSLS, demonstrating superior performance in energy efficiency, delay, prediction error, and localization coverage.