Evaluation of H-infinity Filter in Time Differential Localization Systems
This paper evaluates the performances of H-infinity filter in differential time of arrival (TDoA) localization applications and compares the results with other filters such as extended Kalman filter (EKF) and unscented Kalman filter (UKF) in practical trials. The filters are compared in ideal as well as non-ideal conditions such as different positive and negative values of initial points, presence of erroneous data and excessive noise. The results show that, the H-infinity filter is sub-optimal in ideal conditions as the other filters outperform this filter, but once the initial points are badly selected or data are erroneous, the filter is more robust and accurate than the other variants. Considering the challenging conditions of the indoor environment, use of this filter in localization systems seems to be a good choice.