Improved Circular Correlation Methods for Acquisition in Software GPS Receiver

In this paper the authors explore the Global Positioning System （GPS） signal acquisition and tracking algorithms used in software GPS receiver. Acquisition time is the most important parameter in evaluating the performance of a software GPS receiver in terms of its speed. A trade-off study is done to seek a good balance between the acquisition accuracy and the processing time. The frequency-domain acquisition method by circular correlation, used in a software GPS receiver, has been improved by studying the power spectrum of the Coarse Acquisition （C/A） code alone. The analysis of C/A code reveals that its power spectrum is symmetrical; hence only half the points are required to perform circular correlation. Besides, either half of the spectrum is asymmetrical where a larger amount of power is concentrated in almost one-quarter of the spectrum on its either sides. This further reduces the number of points used to perform correlation. Comparative results of MATLAB simulation of full-size, half-size and quarter-size circular correlation done on actual data stored on hard disk are provided, and they agree with those obtained using GPS receiver. Further reduction in acquisition time has been achieved by investigating the effect of length of the noncoherent pre-integration period. The improved acquisition methods pave way for further development of new algorithms to enhance software GPS receiver performance.

Architecture design of GPS software receiver and implementation of its acquisition algorithm with fine frequency estimation

The design of a global positioning system （GPS） software receiver is introduced. This design uses the concept of software radio, and it consists of the following parts： front-end, acquisition, tracking, synchronization, navigation solution and some assisting modules. In the acquisition module, the acquisition algorithm based on circular correlation is utilized. The input data and the local code are converted into the frequency domain by means of the fast Fourier transform （FFT）. After performing circular correlation, the initial phase of the C/A code can be obtained and the cartier frequency can be found in 1 kHz frequency resolution, which is too coarse to use for the tracking loop. In order to improve the frequency resolution, the fine frequency estimation through a phase relationship is then achieved, by which, the frequency resolution is improved dramatically. Experiments show that the inaccuracy of the carrier frequency can be estimated within a few hertz by the fine frequency estimation method, and the fine frequency attained can be directly used for the tracking loop.