TOA estimation using UWB with low sampling rate and clock drift calibration

In this paper, a time-of-arrival (TOA) estimation scheme with clock drift calibration is proposed using an impulse-radio (IR) ultra-wideband (UWB) signal. We adopt low-rate stroboscopic sampling, which can achieve an equivalent sampling rate as high as the Nyquist sampling rate. The clock drift is one of the main error sources in TOA estimation for the stroboscopic sampling IR-UWB system, since a long preamble is required to collect sufficient data samples. Therefore, the drift is taken into account in our system model. First, the maximum-likelihood estimate of the drift is computed. Then, we employ the peak selection or the leading edge detection to estimate the TOA using the averaged data samples calibrated for the drift. Simulation results corroborate that the drift calibration significantly reduces the TOA estimation errors due to the drift, and that stroboscopic sampling can achieve the same estimation resolution as Nyquist sampling.