Enhanced Vector Tracking of GNSS Signals based on Space-frequency Diversity Reception
Meiling Wang, Yafeng Li, School of Automation, Beijing Institute of Technology, China; Nagaraj C. Shivaramaiah, University of Colorado at Boulder
In harsh environments (indoor, tunnel, urban canyon, dense forest, etc.), it is difficult for traditional GNSS receivers to provide accurate and reliable Position, Navigation and Timing (PNT) services due to signal blockage, attenuation, and multi-path fading. With the expansion and increasing demand for location-based services, the contradiction between the high degree of user dependence on GNSS and the vulnerability of GNSS receivers in adverse conditions is becoming more and more serious.
Traditional GNSS receivers generally achieve fine synchronization between locally generated and received signals by scalar tracking mode. The scalar tracking performance for each signal only depends on the processing capability of the corresponding channel; once the signal quality of a specific channel degrades, its tracking tends to be interrupted. To improve signal tracking robustness in adverse conditions, various joint processing technologies have been developed [1-3]. Among them, the vector tracking method is arguably one of the most effective methods and is regarded as a next-generation tracking approach for advanced GNSS receivers .
The superior performance of vector tracking is derived from the fact that the PVT (Position, Velocity and Time) information is fed back to tracking channels to achieve aiding of weak signal tracking by strong signals . Therefore, in scenarios where all received signals are fairly attenuated, the overall performance of vector tracking would be degraded consequently. In such conditions, increasing the processing gain of weak signals by extending correlation time is limited by navigation bit transition, receiver oscillator instability, and user dynamics. That is, on the basis of single frequency reception, it is impossible to make the signal processing gain substantially improved in simple time-domain operations. Diversity reception technology has been widely used in communication and radar signal processing in signal fading conditions. In recent years, several researchers have tried to adopt diversity reception strategy to enhance processing gain of weak GNSS signals [6-7], providing new ideas for dealing with challenging operating environments for GNSS receivers.
The objective of this paper is to explore methods for combining advantages of both diversity reception and vector tracking in an effective way. For this end, two wide-band GPS antennas are employed to receive L1 C/A and L2C signals to achieve space-frequency diversity reception. To fully exploit the diversity gain, a signal power combining method in correlator level is proposed for diverse signals transmitted by the same satellite. Combined correlation values are then used to construct non-coherent DLL and differential-power FLL discriminators. In addition to the prompt correlation values, the differential-power FLL discriminator also requires the use of slow and fast correlation values , which inevitably increases the complexity of the receiver design. To solve this problem a phase-rotation technique is presented to generate the slow and fast correlation values using the prompt correlation values. This provides a computationally efficient implementation strategy for the diversity combining and makes the correlator structure compatible with the classical GNSS receiver architecture. Code phase and frequency tracking error estimates are finally employed to steer an extended Kalman filter to close the Vector Delay/Frequency Lock Loops (VDFLL).
Four vector tracking methodologies of different signal receptions, namely, single frequency, space diversity, frequency diversity and space-frequency diversity, are implemented and compared. Experiments with real GPS signals captured by pedestrian under heavy foliage are conducted to verify the effectiveness of the proposed tracking method. The results show that the vector tracking based on space-frequency diversity reception significantly outperforms the other three implementations in term of both signal tracking robustness and positioning accuracy in deep signal attenuation/fading conditions.
 Siddiqui, Bashir Ahmed, et al. "Joint Data-Pilot acquisition and tracking of Galileo E1 Open Service signal." IEEE, 2010:1-7.
 Guo, Wenfei, et al. "Design and Performance Evaluation of a Dual Antenna Joint Carrier Tracking Loop." Sensors 15.10(2015):25399-25415.
 Tabatabaei, Amir, and M. R. Mosavi. "Robust adaptive joint tracking of GNSS signal code phases in urban canyons." Iet Radar Sonar & Navigation 11.6(2017):987-993.
 Lashley, Matthew, D. M. Bevly, and J. Y. Hung. "A valid comparison of vector and scalar tracking loops." Position Location and Navigation Symposium IEEE, 2010:464-474.
 Bhattacharyya, Susmita, and D. Gebre-Egziabher. "Development and Validation of Parametric Models for Vector Tracking Loops." Navigation57.4(2010):275–295.
 Siddakatte, Ranjeeth Kumar, A. Broumandan, and G. Lachapelle. "Enhanced GNSS Signal Tracking in Fading Environments using Frequency Diversity." Navigation 64.4(2017).
 Siddakatte, Ranjeeth Kumar, A. Broumandan, and G. Lachapelle. "Use of diversity techniques for weak GNSS signal tracking in fading environments." 134.134(2014):1-10.
 Curran, James T. "A Comparison of Differential-Power and Differential-Phase FLLs for GNSS Receivers." Ion Gnss 2012 - Nashville- Session D2012:1477-1487.