博士 / 國立成功大學 / 航空太空工程學系碩博士班 / 100 / Nowadays, more than 600 millions of people worldwide rely on the satellite navigation to deliver the Position, Velocity, and Time (PVT) information. Meanwhile, with the developing augmentation system or regional satellite systems, the performance of Global Navigation Satellite System (GNSS) for civil users is extremely improved especially on the error budget from system characteristics. However, unpredictable factors on the ground user environment, such as the low received power signal, or the radio interferences, strongly degrade the performance of GNSS receivers. For the reason, this dissertation proposes solutions of the low received power signal and interference issues via time-frequency analyses.
Time-frequency analysis is chosen because it can characterize signals whose spectral characteristics changes with time. In this dissertation, the time-frequency method based on the Fourier sine spectrum is used to analyze the GNSS signals. This method implements a class of fast and diffusive filter to serve as the trend removal tool, and the Fourier sine spectrum, the modified Gabor transform and the Hilbert transform accomplish the Fourier sine spectrogram. It is capable to represent the information on a spectrogram with small error. In this dissertation, several experiments are conducted to evaluate performance of the current available time-frequency tools, such as the short-time Fourier transform, the Gabor transform, and the Winger-Ville distribution. The performance is probed based on the criteria of the time/frequency resolutions, the computation time, and the capability of the weak signal localization. The experiment results show the effectiveness of the Fourier sine spectrogram in representing the detail information in the resulting spectrogram. Finally, this dissertation implements the Fourier sine spectrogram as a time-frequency tool to analyze the GNSS low received power signals and radio frequency interference issues.
As for the low received power GNSS signal issue, a Signal Existence Verification (SEV) process is proposed to detect and subsequently verify the existence of the acquired low received power GPS signal. The SEV scheme is developed based on the Fourier sine based time-frequency representation. This scheme serves as an additional loop for GNSS receiver without changing the original signal processing algorithms, such as the acquisition loop and tracking loop architectures. The SEV receives the satellite code delay phase and Doppler frequency information from the acquisition loop, and subsequently remove the code effect accordingly. The Fourier sine spectrogram of the codeless signal is generated to capture the characteristic of GNSS carrier signal around the corresponding Doppler frequency. In this dissertation, the live low received GNSS signal is collected to present the effectiveness of the SEV scheme.
On the other hand, an interference detection and excision scheme based on the Fourier sine spectrogram is proposed to deal with the GNSS interference issue as well. This detection and excision scheme is designed in accordance with the characteristics of GNSS signal whose signal power is spread below the ambient noise floor. Accordingly, the interference could be detected directly from the spectrogram of the received GNSS signal. Once any interference is detected on the resulting spectrogram, its time-frequency components will be localized in the interference detection stage. An Infinite Impulse Response (IIR) notch filter is implemented in the interference excision stage to further excise the interference source. In this dissertation, several simulations and experiments are conducted to validate the effectiveness of the proposed SEV and the interference detection and excision schemes. The effectiveness of the interference detection and excision scheme is investigated based on the criterion of the estimated signal-to-noise ratio in GNSS signal processing.
Considering the Communications, Navigation, Surveillance/Air Traffic Management Systems (CNS/ATMS), which is developed based on the GPS technique, will be online in Taipei flight information region. A low-cost real-time interference monitoring system is deployed at the Kaohsiung International Airport surroundings to probe the circumstance of radio frequency interference. The long-term observations are collected and investigated to conclude the statistic result of the interference. In addition, several specific interference events are analyzed via the Fourier sine spectrogram. The time-frequency characteristics of the observed interference could provide the information for interference classification, and further the intention reckoning. In this dissertation, the interference influence on the navigation service is also studied using a GNSS software defined receiver. Currently, the results show that observed interference events are not critical to cause the misleading information in positioning. Accordingly, this study could provide some useful information for localizing the interference source and further and mitigating the effects. Also it could be reference for evaluation of the airport surroundings prior to the CNS/ATMs implementation.
| Identifer | oai:union.ndltd.org:TW/100NCKU5295102 |
| Date | January 2012 |
| Creators | Chih-ChengSun, 孫志成 |
| Contributors | Shau-Shiun Jan, 詹劭勳 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | en_US |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 135 |
Page generated in 0.0016 seconds