Title: Simultaneous Frequency Search with a Randomized Dirichlet Kernel for Fast GPS Signal Acquisition
Author(s): Chun Yang, Andrey. Soloviev, Michael Veth, Erik Blasch
Published in: Proceedings of the 29th International Technical Meeting of The Satellite Division of the Institute of Navigation (ION GNSS+ 2016)
September 12 - 16, 2016
Oregon Convention Center
Portland, Oregon
Pages: 92 - 102
Cite this article: Yang, Chun, Soloviev, Andrey., Veth, Michael, Blasch, Erik, "Simultaneous Frequency Search with a Randomized Dirichlet Kernel for Fast GPS Signal Acquisition," Proceedings of the 29th International Technical Meeting of The Satellite Division of the Institute of Navigation (ION GNSS+ 2016), Portland, Oregon, September 2016, pp. 92-102.
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In
Abstract: Signal acquisition in a GPS receiver aims at quickly obtaining coarse estimates of a GPS signal’s time and frequency parameters so as to initialize the code and carrier tracking loops for subsequent refined signal estimation. Conventional methods divide the time and frequency uncertainty zone of the signal into a grid of search points and then test each and every search point by correlating the incoming signal samples with those of a local replica generated with the parameters thereof. If several grid points can be checked at the same time per correlation, the uncertainty zone can be swept over quickly, leading to a fast acquisition process. In this paper, we present a fast acquisition search technique (FAST) via simultaneous search of allowable frequency errors (SAFE). FAST is based on judicious combining of a number of carrier replicas at selected frequency search points, leading to a combined carrier replica (CCR) and a multi-frequency modulated code replica (MMCR). As such, it can reduce the total test points of MN, where M is the number of frequency bins and N is the number of code lags, into M+N. That is, the method achieves fast acquisition using two linear time searches. Optimality criteria and practical methods (a randomized Dirichlet kernel) to reduce implementation loss of CCR and MMCR are described. Simulation and experimental data processing results are presented to demonstrate the functionality and performance of FAST.