A Design Method of Code Correlation Reference Waveform in GNSS Based on Least-Squares Fitting

Chengtao Xu, Zhe Liu, Xiaomei Tang, Feixue WangSensors 2016 · DOI

TL;DR

This paper introduces a novel method for designing Code Correlation Reference Waveforms (CCRWs) for Global Navigation Satellite Systems (GNSS) using a least-squares fitting approach combined with Singular Value Decomposition (SVD). This technique effectively mitigates multipath errors in modern GNSS signals, particularly the complex Binary Offset Carrier (BOC) signals, by avoiding problematic "false lock" points that plague existing methods.

What Problem Does This Paper Address?

Modern GNSS signals, especially the increasingly prevalent Binary Offset Carrier (BOC) signals used in systems like GPS, Galileo, and BeiDou, offer improved accuracy and spectrum separation. However, these signals are highly susceptible to multipath effects – reflections of the signal off surfaces that arrive at the receiver with a delay, causing significant ranging errors. While techniques like Code Correlation Reference Waveforms (CCRWs) help mitigate these errors for simpler Binary Phase Shift Keying (BPSK) signals, they often fail with BOC signals. The core issue is that BOC signals have multi-peaked auto-correlation functions, which can cause tracking loops to lock onto incorrect "side peaks," leading to inaccurate positioning (false lock). Existing CCRW methods struggle to generate discriminators that are both effective against multipath and robust against these false lock points for BOC signals.

Methods

The authors propose a least-squares approximation method to design CCRW discriminators. The key innovation is the use of truncated Singular Value Decomposition (SVD) to handle the ill-conditioned nature of the coefficient matrix that arises from band-limited signal processing. This SVD-based approach allows for a more robust and reliable fitting of the desired "ideal" S-curve (the discriminator's output characteristic) even when dealing with noisy or imperfect data. The method involves:

  1. Formulating the problem as a least-squares fit: The goal is to find optimal weights for a combination of base strobe waveforms to approximate an ideal S-curve.
  2. Using SVD for regularization: Truncating the singular value decomposition helps to regularize the ill-conditioned matrix, preventing unreliable solutions and improving robustness.
  3. Applying to various GNSS signals: The method is tested on BPSK(1) and several BOC signals (BOC(1,1), BOC(2,1), BOC(6,1), BOC(7,1)) across different bandwidths.

Datasets

Not discussed in the paper. The authors appear to use simulated data based on signal models and parameters rather than real-world datasets.

Key Insights and Ideas

  • SVD is crucial for robust CCRW design: The ill-conditioned nature of the coefficient matrix in band-limited scenarios makes standard least-squares fitting unreliable. SVD provides a mathematically sound way to overcome this, leading to stable and accurate discriminator designs.
  • Avoiding false lock points is paramount for BOC signals: The proposed method explicitly aims to generate discriminators that do not have the problematic zero-crossing points that lead to false locks in BOC signal tracking.
  • Trade-offs exist: While the proposed method excels at avoiding false locks and improving multipath error envelopes, there's a general degradation in tracking jitter compared to some established methods. This highlights the inherent trade-offs in designing robust GNSS receivers.
  • Bandwidth impacts performance: The simulations show that narrower bandwidths can sometimes lead to more easily realized reference waveforms and better performance in certain aspects, though it can also make finding suitable waveforms more challenging for higher-order BOC signals.

Key Results

  • Successful avoidance of false lock points: The proposed method consistently generates discriminators that avoid false lock points for both BPSK and BOC signals, a significant improvement over existing CCRW techniques for BOC.
  • Improved Multipath Error Envelope (MEE): For BPSK and BOC(1,1) signals, the proposed discriminators show a reduced multipath error envelope area, indicating better mitigation of multipath effects. For example, the coherent LS discriminator achieved a minimum envelope area for BPSK(1) that was 95.6% of the non-coherent LS, 47.3% of W2, and 44.9% of the double-delta correlator.
  • Tracking Jitter Performance: Generally, the proposed method shows slightly worse tracking jitter compared to established methods like W2 and double-delta correlators, especially for BPSK. However, for BOC(1,1) signals with limited bandwidth, the proposed method's performance was comparable or better.
  • Coherent vs. Non-coherent: Both coherent and non-coherent versions of the proposed discriminators were evaluated, with varying performance characteristics depending on the signal and bandwidth.

Limitations and Future Work

  • Tracking Jitter Degradation: The paper acknowledges that the proposed method generally leads to worse tracking jitter compared to some existing techniques. This is attributed to the complexity of the LS waveforms impacting tracking precision.
  • Ideal S-curve Approximation: The authors note that the "ideal" S-curve might not be the absolute best target, and relaxing certain constraints (e.g., on the right side of the linear region) could potentially improve performance further, though this is difficult to realize in the LS process and is suggested for future research.
  • Hardware Cost: Achieving smaller strobe widths for better performance would necessitate higher sampling frequencies in the receiver, increasing hardware costs.
  • High-Order BOC Signals: For higher-order BOC signals (like BOC(6,1) and BOC(7,1)), only coherent discriminators could be reliably generated due to the complex auto-correlation functions.
  • Future Work: The paper suggests further research into relaxed S-curve restrictions and exploring the trade-offs between MEE, tracking precision, and false lock avoidance more deeply.

Summary updated

Abstract

The multipath effect is one of the main error sources in the Global Satellite Navigation Systems (GNSSs). The code correlation reference waveform (CCRW) technique is an effective multipath mitigation algorithm for the binary phase shift keying (BPSK) signal. However, it encounters the false lock problem in code tracking, when applied to the binary offset carrier (BOC) signals. A least-squares approximation method of the CCRW design scheme is proposed, utilizing the truncated singular value decomposition method. This algorithm was performed for the BPSK signal, BOC(1,1) signal, BOC(2,1) signal, BOC(6,1) and BOC(7,1) signal. The approximation results of CCRWs were presented. Furthermore, the performances of the approximation results are analyzed in terms of the multipath error envelope and the tracking jitter. The results show that the proposed method can realize coherent and non-coherent CCRW discriminators without false lock points. Generally, there is performance degradation in the tracking jitter, if compared to the CCRW discriminator. However, the performance promotions in the multipath error envelope for the BOC(1,1) and BPSK signals makes the discriminator attractive, and it can be applied to high-order BOC signals.
BibTeX
@inproceedings{xu2016design,
  author = {Chengtao Xu and Zhe Liu and Xiaomei Tang and Feixue Wang},
  title = {{A Design Method of Code Correlation Reference Waveform in GNSS Based on Least-Squares Fitting}},
  booktitle = {Sensors},
  year = {2016},
  doi = {10.3390/s16081194},
}
Figure