Maximum likelihood time of arrival estimation for real-time physical location tracking of 802.11a/g mobile stations in indoor environments

This paper describes a new method of time of arrival (TOA) estimation for OFDM signals in indoor multipath environments for application to location tracking systems based on time difference of arrival. The problem addressed is to estimate the line of sight (LOS) time delay between the transmitter and the receiver when the intervening multipath channel is not known. In our approach, we do not attempt to determine the detailed multipath structure of the channel but, instead, use statistical knowledge of the multipath environment to aid in the estimation of the single parameter of real interest. We separate the gross time delay (LOS time delay) from the fine multipath structure of the channel. We then assume a statistical model of the multipath structure and develop the maximum likelihood estimator of the LOS time delay under the assumed model. Analytical development reduces the estimator to an efficient form in which most of the computational effort is done off line. Simulation results are presented which compare the performance of the new estimator to one which uses the well-known MODE technique for estimating the TOA and the new estimator is shown to compare favorably, especially as the number of paths increases. The simulations also show that the estimator is robust to variations in the assumed statistical model.