Computational Intelligence based algorithm for node localization in Wireless Sensor Networks

Accurate location of target nodes is highly desirable in a Wireless Sensor Network (WSN) as it has a strong impact on overall performance of the WSN. This paper proposes the application of H-Best Particle Swarm Optimization (HPSO) and Biogeography Based Optimization (BBO) algorithms for distributed optimal localization of randomly deployed sensors. The proposed HPSO algorithm is modeled for fast and mature convergence, though previous PSO models had only fast convergence but less mature. Biogeography is a school work (collective learning) of geographical allotment of biological organisms. BBO has a new inclusive vigor based on the science of biogeography and employs migration operator to share information between different habitats, i.e., problem solutions. WSN localization problem is formulated as an NP-Hard optimization problem because of its size and complexity. In this work, an error model is described for estimation of optimal node location in a manner such that the location error is minimized using HPSO and BBO algorithms. Proposed HPSO and BBO algorithms are matured to optimize the sensors' locations and perform better as compared to the existing optimization algorithms such as Genetic Algorithms (GAs), and Simulated Annealing Algorithm (SAA). Comparative study reveals that the HPSO yields improved performance in terms of faster, matured, and accurate localization as compared to global best (gbest) PSO.