Genetic Algorithm for Node Localization in WSN

Dr. G. U. Kharat, R. S. Bansode, H. A. Chavan

Abstract: Wireless Sensor Networks (WSNs) often face challenges in accurately determining node locations due to the limitations of traditional localization methods like GPS, which can be prohibitively expensive for large networks. Moreover, while range- based techniques like Angle of Arrival (AoA), Time of Arrival (ToA), and Time Difference of Arrival (TDoA) offer improved accuracy, they require additional hardware, adding complexity and cost. This study aims to implement a hybrid localization approach leveraging Received Signal Strength (RSS)- based measurements, optimized using a Genetic Algorithm (GA), for localizing sensor nodes within a WSN. By harnessing RSS values from anchor nodes, the proposed method estimates node positions and minimizes errors in location determination. The use of GA allows for effective exploration and exploitation of the search space, enabling the algorithm to converge to an optimal solution efficiently. Through the application of genetic operators such as selection, crossover, and mutation, the GA refines the solution set iteratively, ensuring convergence towards the minimum localization error. The method also mitigates issues like multipath interference and noise in signal strength, improving the robustness of the system. The proposed approach is modelled and tested in MATLAB, where the simulation results show that the hybrid GA- RSS algorithm significantly enhances localization accuracy while maintaining low computational complexity. The method achieves precise node localization without incurring the high costs and hardware requirements of traditional approaches, making it particularly suitable for large- scale WSN deployments in cost-sensitive applications like environmental monitoring, smart cities, and military surveillance. This work provides a scalable and efficient solution to the node localization problem, optimizing resource usage while maintaining the accuracy necessary for reliable data gathering and communication within WSNs.

Keywords: wireless sensor networks, hybrid localization, received signal strength, genetic algorithm, node positioning