Mitigation of sampling-induced errors in delay estimation

Delay estimation has been well accepted as an important approach to ranging with a UWB (ultra-wide band) system. Delay estimation errors due to digital sampling can be considerably large at a low sampling frequency. In this paper, we investigate reduction of such errors in both analytical and practical aspects. A delay estimation scheme based on the so-called simplified maximum likelihood formulation is first derived, and is further shown to be optimal in the sense that its error variance converges to the corresponding theoretical lower limit as the sampling frequency is sufficiently high. Next, an iterative scheme for first arrival detection is devised in light of the near-far robustness in multiuser detection. This scheme can effectively cope with a frequent UWB environment where the first arriving signal component may not be dominant among multipath components. Efficient reduction of sampling-induced errors and a low computation complexity are main advantages of the proposed schemes.