An Efficient NLOS Errors Mitigation Algorithm for TOA-Based Localization.

Area of Science:

• Signal Processing
• Localization Systems
• Optimization Theory

Background:

• Non-line-of-sight (NLOS) propagation significantly impacts time-of-arrival (TOA) localization accuracy.
• Existing NLOS error mitigation methods often rely on unavailable information like error statistics or noise variances.
• Practical TOA localization demands robust methods that do not assume prior knowledge of error characteristics.

Purpose of the Study:

• To develop a novel NLOS error mitigation scheme for TOA localization systems.
• To address the practical limitation of unknown NLOS error statistics and TOA measurement noise variances.
• To extend the proposed scheme for cooperative source localization scenarios.

Main Methods:

• Utilizes a constrained least-squares optimization approach.
• Transforms the optimization problem into a semidefinite programming (SDP) formulation.
• Employs the CVX toolbox for efficient problem solving.
• Extends the core algorithm for cooperative localization.

Main Results:

• The proposed scheme effectively mitigates NLOS errors without requiring prior statistical information.
• Performance evaluation through extensive simulations demonstrates superior accuracy compared to existing methods in most scenarios.
• The SDP-based approach provides a computationally tractable solution for NLOS mitigation.

Conclusions:

• The developed NLOS error mitigation technique enhances TOA localization accuracy in practical, information-scarce environments.
• The semidefinite programming formulation offers a robust and efficient solution for NLOS error reduction.
• The extended cooperative localization scheme shows significant performance improvements, validating the method's effectiveness.