Block-Repetitive Iterated Processing For Software Gps Receiver: Dichotomized Search Of Correlation Peak

The software GPS receiver offers a block-repetitive iterated processing capability that is difficult, if not impossible, to realize with a hardware correlator-based GPS receiver. Applications of such block-repetitive iterated processing include iterative estimation of GPS signal parameters under high dynamics and nonlinearities, cross-correlation cancellation of near-far problem and multi-access interference, and multipath mitigation.In this paper, we first analyze the block-repetitive iterated processing capability offered by a software GPS receiver in both the acquisition and tracking modes as compared to conventional hardware correlator-based GPS receivers. Then we present a dichotomized search of the true correlation peak for delay estimation as an illustrating example of block iterative processing. This is in contrast to a conventional delay-locked loop (DLL) that moves several correlators of fixed spacing so as to drive the delay error discriminator to zero (an early-minus-late type). An advantage of this dichotomized search is its reduced sensitivity to multipath. This can be understood as if it implements a sequential software narrow correlator vs. parallel hardware narrow correlators. Since it adapts to data with a variable spacing, it is computationally more efficient than a software implementation that uses a fixed number of correlators with the same coverage.Simulation results are presented to show the multipath-desensitized delay estimation as compared to the theoretical multipath error envelope and to the thermal noise errors under various SNR conditions for different multipath parameters.