Prior Models for Indoor Super-Resolution Time of Arrival Estimation

We propose a scheme for super-resolution estimation of the time of arrival of a radio signal which makes use of a prior model of the radio propagation conditions. The method is based on examination of the leading edge of the channel impulse response, and comparison with a database of leading edge shapes generated by Monte Carlo simulation. We also propose an efficient implementation, which makes the method practical for mobile devices with limited computational power. We verify the performance of the method using actual measured data, taken in a number of different situations, and show improved performance over other super-resolution methods proposed in the literature.