Distributed RSS-based localization in wireless sensor networks based on second-order cone programming.

Area of Science:

• Computer Science
• Electrical Engineering
• Wireless Communication

Background:

• Received Signal Strength (RSS)-based localization is crucial for Wireless Sensor Networks (WSNs).
• Existing cooperative localization methods face challenges in accuracy and computational efficiency.
• Maximum Likelihood (ML) criterion offers asymptotically efficient solutions but often involves non-convex optimization.

Purpose of the Study:

• To propose a novel convex optimization approach for RSS-based cooperative localization in WSNs.
• To enhance localization accuracy and reduce energy consumption.
• To address the non-convexity inherent in ML-based localization problems.

Main Methods:

• Formulation of the localization problem using the Maximum Likelihood (ML) criterion.
• Application of convex relaxation techniques to transform the non-convex ML problem into Second-Order Cone Programming (SOCP).
• Development of an iterative, local positioning scheme using adjacent node measurements and a heuristic for convergence improvement.

Main Results:

• The proposed convex optimization approach achieves estimation accuracy improvements exceeding 1.5 m over existing methods.
• The new method requires fewer iterations for convergence, leading to reduced energy consumption.
• Demonstrated superior performance in terms of accuracy and energy efficiency across various scenarios.

Conclusions:

• The convex optimization-based approach provides a more accurate and energy-efficient solution for RSS-based cooperative localization in WSNs.
• The SOCP formulation effectively handles the non-convexity of the ML optimization problem.
• The proposed method offers a significant advancement over current state-of-the-art localization techniques in WSNs.