Elephant Herding Optimization for Energy-Based Localization.

Area of Science:

• Wireless Sensor Networks
• Optimization Algorithms
• Signal Processing

Background:

• Source localization is crucial for wireless sensor networks (WSNs).
• Traditional methods often rely on approximations or convex relaxations for the maximum likelihood (ML) problem.
• Directly solving the ML problem using metaheuristics has not been explored for WSNs.

Purpose of the Study:

• To apply metaheuristics, specifically the Elephant Herding Optimization (EHO) algorithm, to solve the energy-based source localization problem in WSNs.
• To directly address the maximum likelihood (ML) problem without approximations.
• To evaluate the performance of the EHO algorithm against existing non-metaheuristic methods.

Main Methods:

• Implementation of the Elephant Herding Optimization (EHO) algorithm for source localization.
• Optimization of EHO algorithm parameters to match the energy decay model in WSNs.
• Comparative analysis of computational complexity and performance against non-metaheuristic algorithms.

Main Results:

• The EHO algorithm demonstrates superior performance in noisy environments compared to existing solutions.
• Optimized EHO parameters effectively model energy decay between sensor nodes.
• The proposed metaheuristic approach provides a viable alternative for direct ML problem-solving.

Conclusions:

• Metaheuristics, particularly EHO, offer a powerful and effective approach for energy-based source localization in WSNs.
• The EHO algorithm shows significant advantages in challenging, noisy network conditions.
• This work encourages further research into metaheuristic applications for WSN localization problems.