GNSS Receiver Jamming Estimation Using Lightweight Logistic Regression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing GNSS receivers face challenges in timely and lightweight jamming estimation, which affects the reliability of positioning information and requires effective countermeasures.
Innovation Solution
A GNSS receiver uses band-specific parameters measured for RF and IF bands, input into a state-less and lightweight logistic regression machine learning model, to estimate the likelihood of jamming, enabling timely countermeasures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If conventional jamming detection methods are used, then the system can detect jamming, but the detection is not timely and requires heavy computational resources
Solution Approach 1:
The patent transforms the jamming detection problem by changing the parameters being measured from complex signal analysis to simple statistical parameters (mean, variance, skewness, kurtosis) of the received signal. This parameter transformation enables timely detection using lightweight computations while maintaining detection effectiveness through the trained logistic regression model.
Solution Approach 2:
The patent replaces complex mechanical signal processing systems with a machine learning-based statistical analysis system. Instead of using heavy computational algorithms for jamming detection, the system uses a trained logistic regression model that processes simple statistical parameters, significantly reducing computational complexity while maintaining detection accuracy.
2Measurement precision
If complex machine learning models are used for jamming estimation, then the estimation accuracy improves, but the model cannot run on receivers with constrained power and computation resources
Solution Approach 1:
The patent employs a lightweight logistic regression model that can be deployed on resource-constrained devices. This simplified model sacrifices some complexity but maintains sufficient accuracy for jamming detection while consuming minimal power and computational resources, making it suitable for embedded GNSS receivers.
Solution Approach 2:
The patent changes the approach from using complex model parameters to simple statistical parameters (mean, variance, skewness, kurtosis) that can be computed efficiently. This parameter simplification enables the use of lightweight machine learning models that run on battery-powered devices without significant power consumption.
3Measurement precision
If band-specific parameters are collected for RF and IF bands, then the jamming estimation becomes more precise, but the parameter collection and processing becomes more complex
Solution Approach 1:
The patent segments the jamming detection process by band (RF and IF bands), collecting specific statistical parameters for each band. This segmentation allows the system to identify jamming in specific frequency bands independently, improving precision while keeping the processing complexity manageable through the use of simple statistical measures for each segment.
Solution Approach 2:
The patent creates a universal parameter collection framework that works across different bands (RF and IF) using the same statistical measures (mean, variance, skewness, kurtosis). This universal approach simplifies processing by applying the same methodology across multiple bands rather than requiring band-specific complex processing algorithms.
Data Source
AI summary
A method for estimating jamming in a global navigation satellite system, GNSS, receiver is provided. The method is performed in the receiver and comprises receiving GNSS signals at a radio frequency band and processing the received signals at an intermediate frequency band; collecting a set of parameters at the receiver based on the received GNSS signals; and obtaining a likelihood value using a machine learning model trained for the receiver, wherein the set of parameters are inputs to the machine learning model, the likelihood value is an output of the machine learning model, and the likelihood value is a number between 0 and 1 and corresponds to a likelihood of the receiver being jammed in the intermediate frequency band.