GNSS Signal Distortion Detection via Autoregressive Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing GNSS systems face challenges in accurately detecting signal deformations caused by faults in satellite transmitter equipment, which can lead to positioning errors, and current methods are sensitive to noise and fail to differentiate between transmission errors and multipath errors.
Innovation Solution
A method involving autoregressive parametric modeling and linear prediction error calculation is used to detect signal deformations, with steps including determining an autoregressive model, calculating linear prediction errors, and comparing them to a detection threshold to alert for signal nonconformity, thereby invalidating pseudo-distance measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If correlation function-based detection is used, then transmission errors can be detected, but false alarms increase due to inability to differentiate from multipath errors and sensitivity to ambient noise
Solution Approach 1:
The detection process is segmented into distinct stages: first computing the correlation function, then applying smoothing filtering to separate the signal components, and finally detecting transmission errors only in the smoothed portion. This segmentation allows differentiation between multipath errors (affecting raw correlation) and transmission errors (affecting smoothed correlation), reducing false alarms.
Solution Approach 2:
Smoothing filtering is applied as a preliminary action before error detection. By pre-processing the correlation function output through smoothing, the method prepares the signal in advance to eliminate multipath interference and noise, ensuring that subsequent error detection operates on cleaned data and reduces false positive rates.
2Object-affected harmful factors
If smoothing filtering is applied to correlation output, then multipath errors are reduced, but detection sensitivity to transmission errors must be maintained
Solution Approach 1:
The method applies smoothing filtering partially - only to the extent necessary to remove multipath errors while preserving transmission error characteristics. By controlling the smoothing degree and applying it selectively to specific portions of the correlation output, the method achieves sufficient multipath rejection without over-smoothing that would mask transmission errors.
Solution Approach 2:
Different portions of the correlation output are treated differently: the smoothed portion is used for transmission error detection while acknowledging that unsmoothed portions may contain multipath effects. This local differentiation in processing quality allows the system to extract transmission error information from smoothed regions while being aware of multipath presence in other regions.
Data Source
Figure 1~2
AI summary
Method for detecting the distortion of a GNSS signal transmitted by at least one GNSS satellite and received by at least one GNSS receiver, said distortion originating from a defect in the generation of the GNSS signal, said method being characterized in that it comprises at least the following steps: o a step of determining at least one autoregressive parametric model of the GNSS signal at the output of a correlation stage comprising said GNSS receiver, o a step of calculating at least one linear prediction error e(n) between said output signal of the correlation stage and said autoregressive parametric model, o at least one step of comparing the linear prediction error to a detection threshold, o a step of deciding on the distortion of the transmitted GNSS signal in the case where the linear prediction error exceeds said detection threshold.