A multi-step Gaussian filtering approach to reduce the effect of non-Gaussian distribution in aerial localization of an RF source in NLOS condition

• Published in: 2013 First RSI/ISM International Conference on Robotics and Mechatronics (ICRoM) pp. 43 - 48
• Main Authors: Dehghan, S. M. M., Moradi, H.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.02.2013
• Subjects: Atmospheric measurements; Equations; Extended Kalman Filter; Localization; Measurement uncertainty; NLOS propagation; Particle Filter; Particle filters; Particle measurements; Scattering; Shadow mapping; Unscented Kalman Filter
• ISBN: 1467358096; 9781467358095
• DOI: 10.1109/ICRoM.2013.6510079

The hybrid localization using Angle of Arrival (AOA) and Received Strength Signal Indicator (RSSI) of an RF source, such as a cell phone in a search and rescue mission, with unknown power and None Line Of Sight (NLOS) condition have been proven to be advantageous compared to using each method separately. The hybrid approach has been proposed to benefit from both RSSI and AOA measurements. In this paper, the initial hybrid method, which was implemented using particle filters due to the multi-modal/non-Gaussian nature of localization in NLOS condition, has been replaced by a multi-step Gaussian filtering approach which provides nearly similar accuracy with better performance. The proposed method has been implemented using extended Kalman filter and Unscented Kalman filter. The simulation results show that the multi-step Gaussian filtering is comparable to particle filter in all cases with better performance. For further evaluation, the effects of uncertainty in the propagation parameters have been studied to show the robustness of each filter to these uncertainties.