GNSS spoofing detection method and device based on LSTM-transformer model, equipment and medium
By constructing a multicorrelator structure in the GNSS receiver tracking loop and combining LSTM and Transformer models to capture the autocorrelation function distortion characteristics, the accuracy and generalization of GNSS spoofing detection are improved, solving the problem of insufficient detection accuracy and generalization in existing technologies.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 湖南工商大学
- Filing Date
- 2026-02-12
- Publication Date
- 2026-06-12
AI Technical Summary
In existing GNSS spoofing detection methods, the traditional three-correlator structure is not sufficient to capture the autocorrelation function distortion under the influence of spoofing signals, resulting in low utilization of detection information. Furthermore, deep learning solutions often rely on a single model and do not fully combine the advantages of different models, leading to insufficient detection accuracy and generalization in complex scenarios.
A multi-correlator structure is constructed in the tracking loop of a GNSS receiver to obtain the correlator output sequences under multiple code phase offsets. The sequences are then processed using an LSTM-Transformer dual-branch timing model. By combining the advantages of LSTM and Transformer branches, local and global timing features are captured. Spoofing detection is achieved through a cross-attention fusion module and a spoofing detection classification module.
It improves the accuracy and generalization of GNSS spoofing detection in complex scenarios, enhances the ability to detect spoofing signals, and is significantly superior to single-model schemes.
Smart Images
Figure CN121679627B_ABST