Majorization-Minimization based Hybrid Localization Method for High Precision Localization in Wireless Sensor Networks

Kuntal Panwar,Prabhu Babu
2023-11-28
Abstract:This paper investigates the hybrid source localization problem using the four radio measurements - time of arrival (TOA), time difference of arrival (TDOA), received signal strength (RSS), and angle of arrival (AOA). First, after invoking tractable approximations in the RSS and AOA models, the maximum likelihood estimation (MLE) problem for the hybrid TOA-TDOA-RSS-AOA data model is derived. Then a weighted least-squares problem is formulated from the MLE, which is solved using the principle of the majorization-minimization (MM), resulting in an iterative algorithm with guaranteed convergence. The key feature of the proposed method is that it provides a unified framework where localization using any possible merger out of these four measurements can be implemented as per the requirement/application. Extensive numerical simulations are conducted to study the performance of the proposed method. The obtained results indicate that the hybrid localization model improves the localization accuracy compared to the heterogeneous measurements under different network scenarios, which also includes the presence of non-line of sight (NLOS) errors.
Signal Processing
What problem does this paper attempt to address?