Accurate time delay estimation based on SINC filtering

Abstract

Time delay estimation is very important in many applications such as radar, sonar, communication, geological exploration, biomedical engineering, etc. We propose a SINC filtering algorithm (SFA), i.e., one based on the sinc function. The algorithm can get an accurate time delay estimate with little or no prior knowledge about the statistical characteristics of the received signals. SFA is based on the theory that time delay estimation can be transformed to the parameter estimation of a SINC filter, and the time delay estimate can be obtained when the maximum of the filter coefficients is known. When the time delay of two received signals is formulated, the LMS algorithm can be used to get the estimate of the SINC filter coefficients, then the time delay estimate can be obtained. Compared with the traditional parabolic fitting method, the SFA can get more accurate results without increasing the computation burden. Computer simulations and water tank experiments both indicate that the SFA can obtain accurate time delay estimation.