A Real-Time Interference Monitoring Technique for GNSS Based on a Twin Support Vector Machine Method

Area of Science:

• Signal Processing
• Machine Learning
• Navigation Systems

Background:

• Global Navigation Satellite System (GNSS) performance is significantly degraded by signal interference.
• Effective interference monitoring is crucial for robust GNSS anti-interference strategies.
• Interference monitoring is framed as a classification problem for advanced signal analysis.

Purpose of the Study:

• To propose a real-time interference monitoring technique for GNSS signals.
• To develop and validate a Twin Support Vector Machine (TWSVM) model for interference detection.
• To assess the efficiency and practical applicability of the TWSVM-based approach.

Main Methods:

• A Twin Support Vector Machine (TWSVM) model was established and solved using the Least Squares Twin Support Vector Machine (LSTWSVM) algorithm.
• Interference monitoring indicators were analyzed to extract features from interfered GNSS signals.
• The proposed method was tested using GPS L1 C/A code signals and compared against standard Support Vector Machine (SVM).

Main Results:

• Experimental results confirmed that selected observations serve as effective interference monitoring indicators.
• The TWSVM-based interference monitoring demonstrated significantly higher speed compared to conventional SVM.
• TWSVM achieved millisecond-level training times and microsecond-level monitoring times.

Conclusions:

• The proposed TWSVM-based interference monitoring technique is highly effective for GNSS signals.
• The method's speed and efficiency make it suitable for practical, real-time interference monitoring applications.
• This approach enhances the reliability and performance of GNSS receivers in the presence of interference.