Adaptive Carrier Smoothing Using Code and Carrier Divergence

摘要

This paper investigates an adaptive carrier smoothing for a single frequency GPS user. Using a Hatch filter in a single frequency receiver reduces multipath and receiver noise but introduces an induced bias that depends on the rate of ionospheric delay and a carrier smoothing time. Ideally, to use a Hatch filter more effectively and efficiently, the rate of ionospheric delay and noise characteristics should be known. However, since these are usually not known to a single frequency user, a conservative constant carrier smoothing time is typically used for safety of life applications such as WAAS and LAAS. rnThis paper starts by introducing how to estimate the rate of ionospheric delay with white noise-like multipath and highly correlated multipath from code and carrier divergence (CCD). For the different multipath characteristics, the ordinary least square (OLS) or the (feasible) generalized least square (GLS) estimators is used. Given the estimated ionospheric delay rate, multipath can be separated and is modeled as white noise or a cosine wave according to the multipath characteristics. The estimated rate of ionospheric delay and the modeled multipath are used to construct a cost function. From the cost function, it is possible to compute an optimal carrier smoothing time which balances the reduction of multipath against the magnitude of the induced bias. These processes are repeated every 60 seconds and update the optimal carrier smoothing time by keep observing the ionospheric rate and multipath. Preliminary results show that this technique significantly suppresses multipath and does not introduce any noticeable large drifts.