Session A2: GNSS Security: Interference, Jamming, and Spoofing 1

Same Frequency Interference and Suppression of GBAS VDB and VOR/ILS Based on the Fast ICA Algorithm
Ziqi He, Hongxia Wang, Kun Fang, Beihang University; Xiao Li, China Satellite Network Group Co., Ltd; Zhipeng Wang, Beihang University
Location: Beacon A

The ground-based navigation system provides aircraft with position, orientation, and distance information by transmitting signals from the ground, aiding in the safe flight and landing of aircraft. Among them, the Ground-Based Augmentation Systems VHF Data Broadcast (GBAS VDB) equipment is an essential component of the airborne GBAS ground station. It integrates with the onboard GBAS positioning and navigation equipment to guide aircraft during precision approach and landing. The Instrument Landing System (ILS) provides precise course and glide slope information during aircraft landing, ensuring a safe touchdown. The VHF Omnidirectional Range (VOR) system offers azimuth information between the aircraft and ground beacons, enabling accurate position determination and route guidance in aviation navigation. The operational stability and reliability of these three systems are critical for maintaining precise flight paths during the approach and landing phases, playing a key role in aviation safety.
One of the main challenges facing civil aviation ground-based navigation systems is interference between systems operating in the same frequency band. GBAS VDB, ILS, and VOR all operate within the 108.000-117.975 MHz frequency range. Issues such as poor airport frequency planning, insufficient geographical separation between systems, frequency misapplication, and equipment malfunctions can lead to interference, negatively impacting system performance. In 2018, the International Civil Aviation Organization (ICAO) Spectrum Working Group (SWG) initiated preliminary updates and coordination efforts to improve the frequency compatibility of GBAS VDB with ILS localizers and VOR transmitters. By 2021, ICAO revised its Manual on Radio Spectrum Requirements, proposing general methods for analyzing the compatibility of ILS, VOR, and GBAS VDB. While these steps laid a foundation for addressing interference, there remains a need for further research into specific technical methods and algorithms to effectively mitigate interference.
Although much of the current research focuses on interference analysis from broadcast signals affecting ground-based navigation systems, there is limited work on suppressing same-frequency interference between GBAS VDB, ILS, and VOR signals. Blind Signal Separation (BSS) techniques, which have gained prominence in signal processing, offer a promising approach to separating mixed signals. The foundational work on BSS was conducted by Jutten C. and Herault J. in 1985. In 1999, Hyvarinen A. proposed the Fast Independent Component Analysis (Fast ICA) algorithm, also known as the fixed-point algorithm, which uses negentropy to assess the Gaussianity of signals, allowing for more efficient separation. In recent years, blind signal separation has shown some progress in mitigating same-frequency interference in navigation signals. For instance, in 2019, Ni et al. proposed a method using blind signal separation to suppress same-frequency interference in airborne VDB receivers, though their work focused mainly on separating same-frequency VDB interference signals. In 2023, He et al. applied machine learning techniques to separate blind source signals. However, no research has yet explored the use of blind signal separation to suppress same-frequency interference between GBAS VDB, VOR, and ILS systems.
Building upon this background, our study first analyzes the mutual interference between same-frequency GBAS VDB, VOR, and ILS signals. We propose a practical compatibility testing scheme, which involves selecting the appropriate equipment, setting up the testing platform, conducting preliminary tests, collecting data, and performing statistical analyses. This testing scheme was verified to be feasible and reasonable. Subsequently, we applied the Fast ICA algorithm, a widely used method in blind source separation, to separate ILS and VOR interference signals from VDB reception signals, aiming to efficiently isolate same-frequency interference signals and extract the desired target signals.
This study mainly includes the following three parts:
The first phase of the study involves developing a robust compatibility testing scheme for GBAS VDB, VOR, and ILS signals. This scheme is based on both domestic and international research on the compatibility of these systems, particularly in scenarios involving the same-frequency operation. Preliminary research and testing were conducted using an IFR-4000 aviation standard signal generator, along with a GBAS VDB transmitter. Signal parameters were meticulously configured to generate VOR, ILS, and VDB signals for analysis. Data collection was performed through a Multi-Mode Receiver (MMR), which is equipped with built-in modules for ILS and VOR, enabling the provision of precise localizer and accurate flight azimuth information. To assess the impact of interference, the failure criteria of aeronautical radionavigation services were applied. An evaluation method for measuring and analyzing the collected data was devised, which provided clear insights into how same-frequency mixed signals from VDB, VOR, and ILS interfere with one another. The experimental results demonstrated that this interference is severe, significantly degrading the performance of reception equipment. This underscores the urgent need for effective interference suppression techniques.
Secondly, given the critical issue of severe same-frequency interference between GBAS VDB, VOR, and ILS, the second phase of the study proposes an improved Fast ICA algorithm. This algorithm leverages negentropy to measure the Gaussianity of signals, facilitating the separation of mixed signals, specifically targeting the separation of same-frequency VOR and ILS interference signals from the VDB signals received by airborne VDB receivers. The algorithm was carefully tailored to the different signal characteristics of VOR and ILS, with the contrast function G(u) selected based on the statistical properties of the signals. The mixed signals from VDB and either VOR or ILS were then incorporated into an initial mixing matrix, with channel noise added to simulate real-world interference conditions. The resulting mixed model was whitened to eliminate the correlation between the mixed signals, reducing the dependency among them. Additionally, to further enhance efficiency, the traditional Fast ICA algorithm using Newton’s iteration method was modified to employ a third-order convergence iteration method. This significant improvement drastically accelerates convergence, allowing for faster and more efficient extraction of the target GBAS VDB signals. The result is an effective suppression of same-frequency interference from VOR and ILS signals, ensuring clearer and more reliable signal reception.
The final phase of the study involved verifying the performance of the proposed method through a series of comprehensive simulations. To gauge the effectiveness of the improved Fast ICA algorithm, its convergence speed and separation performance were compared against the basic Fast ICA algorithm. A similarity coefficient matrix was employed to analyze the degree of signal separation achieved. The simulation results provided compelling evidence that the improved Fast ICA algorithm significantly outperforms its predecessor, both in terms of faster convergence and superior signal separation. Not only did the improved algorithm excel in isolating VOR and ILS interference from VDB signals, but it also led to a substantial improvement in overall signal reception quality. This validation reaffirms the feasibility and practicality of the proposed method for real-world applications, offering a promising solution to the persistent problem of same-frequency interference in aviation navigation systems.