A Modified Particle Filter for Cooperative Positioning

This paper proposes a modified hybrid cooperative particle filter (MHC-PF) for cooperative positioning in Global Positioning System (GPS)-challenged scenarios, utilizing information from both satellites and terrestrial neighboring GPS receivers. In GPS-challenged scenarios, determination of receivers' positions is still a challenging task due to radio blockage. In this situation, cooperative positioning can be utilized to improve the ability to estimate position. The proposed MHC-PF involves introducing a modified factor to the likelihood function, and then selecting a value of the modified factor that results in a minimum estimation error through Monte-Carlo strategy in a pre-processing stage. The proposed method is verified by a realistic indoor scenario to demonstrate the accuracy and availability. Simulation results indicate that the proposed MHC-PF provides approximately 2-m horizontal position root mean squared error (RMSE) and significant improvements over the existing method.