BeiDou positioning signal interference recognition and processing algorithm based on time-frequency analysis
In controlled BeiDou B1I simulation scenarios, traditional recognition and suppression algorithms may show limited adaptability when interference types and jamming-to-noise ratios vary. Based on time-frequency analysis, this paper studies a two-stage BeiDou positioning signal interference recognition and processing algorithm. First, signal models for six representative interference types are constructed, and their time-frequency characteristics are extracted. Second, a CNN-based recognition model is used to classify the time-frequency feature maps, while WPT is adopted for multi-scale interference suppression and signal reconstruction. The contribution of this study lies in integrating feature-guided CNN recognition with WPT-based suppression in a unified and computationally efficient pipeline for BeiDou B1I interference simulation. Experiments on independently generated simulated samples show that the CNN model achieves an average recognition accuracy of 94.01%, a false positive rate of 3.31%, and a recognition time of 3.25 s for six interference categories. The WPT-based suppression stage obtains an MSE of 2.54%, a signal recovery rate of 94.63%, an SNR of 28.44 dB, and a calculation time of 0.18 s. These results indicate that the proposed pipeline is effective under the tested simulation conditions; field-measured BeiDou/GNSS data and more complex composite interference should be further evaluated before extending the conclusions to practical electromagnetic environments.
Authors
- Xiao Feng
- Fang Yuan (ORCID: https://orcid.org/0009-0008-6813-3631)
- Yong Liu
- Yao Zhao
Institutions
- State Grid Corporation of China (China) (CN)
- Henan Energy & Chemical Industry Group (China) (CN)
- Shanghai Electric (China) (CN)
Publication Details
- Journal
- Discover Computing
- Published
- 2026-11-14
- DOI
- https://doi.org/10.1007/s10791-026-10717-6
- Primary Topic
- GNSS positioning and interference
- Type
- article
- Field-Weighted Citation Impact
- 0.00