Efficient super resolution time delay estimation techniques

In this paper, an e cient Weighted Fourier transform and RELAXation based algorithm (referred to as WRELAX) is rst proposed for the well-known time delay estimation problem. WRELAX involves only a sequence of weighted Fourier transforms. Its resolution is much higher than that of the conventional matched lter approach. One disadvantage associated with WRELAX is that it converges slowly when the signals are spaced very closely. To overcome this problem, the well-known high resolution MODE (Method of Direction Estimation) algorithm, which was originally proposed for angle estimation in array processing, is modi ed and used with WRELAX for super resolution time delay estimation. The latter method is referred to as MODE-WRELAX. MODE-WRELAX provides better accuracy than MODE and higher resolution than WRELAX. Moreover, it applies to both complexand real-valued signals (including those with highly oscillatory correlation functions). Numerical results show that the MODE-WRELAX estimates can approach the corresponding the Cram er-Rao bounds.