GPS positioning in a multipath environment

We address the problem of GPS signal delay estimation in a multipath environment with a low-complexity constraint. After recalling the usual early-late estimator and its bias in a multipath propagation context, we study the maximum-likelihood estimator (MLE) based on a signal model including the parametric contribution of reflected components. It results in an efficient algorithm using the existing architecture, which is also very simple and cheap to implement. Simulations show that the results of the proposed algorithm, in a multipath environment, are similar to these of the early-late in a single-path environment. The performance are further characterized, for both MLEs (based on the single-path and multipath propagation) in terms of bias and standard deviation. The expressions of the corresponding Cramer-Rao (CR) bounds are derived in both cases to show the good performance of the estimators when unbiased.