Design and Analysis of a Hot Start Vector Tracking Receiver

Abstract

This thesis presents a novel method of initializing vector processing on a software-defined Global Positioning System receiver. By leveraging observables and ephemeris data obtained from a base station, the proposed receiver is able to instantly begin vector tracking. Conventional vector tracking approaches require an initial position estimate and decoded ephemeris data to commence processing, which is typically provided by scalar tracking. The proposed method overcomes this limitation and generates an initial position estimate without prior knowledge of the signal or receiver states.

The receiver architecture integrates additional correlators located beyond the conventional half-chip range of the GPS L1 replica spacing vector. These extended correlators provide key insight into the signal power distribution, allowing the tracking loop to accurately align the code replica even when the initial position estimate is inadequate or the code phase lies outside the typical Delay Lock Loop pull-in region. This capability is particularly advantageous in dynamic or degraded environments where robust tracking is essential.

By combining base station assisted initialization with extended correlator design, the proposed approach effectively merges the rapid time-to-first-fix of hot start methods with the robustness of vector tracking. The resulting system provides a computationally efficient, practical, and reliable solution for GPS receivers operating in a variety of signal conditions. In this thesis, the HSVT algorithm results show state convergence in under 1 second for static receivers and under 2 seconds for dynamic receivers, demonstrating performance comparable to existing hot start methods. These convergence times are validated through both experimental live-sky scenarios and simulated Monte Carlo analyses.