Machine Learning Module for Jammed Frequency Slot Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing jamming detection methods in frequency hopping signals are limited by the need for specific scenario design and pre-defined characteristics, making them inefficient for detecting jammers and requiring high design efforts, with a limited field of application.
Innovation Solution
A method for training a machine learning module using IQ samples from jammed frequency hopping signals, where the module executes an artificial neural network to classify jammed and benign hops or slots, allowing for robust detection of jammers without pre-defined features and adaptable to various scenarios through supervised or semi-supervised learning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If algorithms and heuristics are designed specifically for a certain field of application with pre-defined characteristics, then detection precision for known scenarios is improved, but adaptability to different jammer scenarios deteriorates
Solution Approach 1:
The system uses unsupervised learning algorithms that automatically adapt to different jammer scenarios without requiring manual reconfiguration or pre-defined characteristics. The machine learning module self-adjusts its detection parameters based on the received signal characteristics, enabling it to handle various jammer types (noise jammers, tone jammers, spread spectrum jammers) with the same algorithmic framework.
Solution Approach 2:
The invention changes the fundamental parameter of detection from fixed pre-defined characteristics to dynamically learned characteristics. By using unsupervised learning, the system automatically adjusts its detection parameters based on the statistical properties of the received signal, allowing it to adapt to different jammer scenarios without manual intervention.
2Measurement precision
If pre-defined characteristics and scenario design are used for jamming detection, then detection accuracy for anticipated scenarios is improved, but design complexity and effort increase
Solution Approach 1:
The system eliminates the need for manual scenario design and characteristic definition by employing unsupervised learning algorithms that automatically learn the relevant features from the received signals. The machine learning module performs self-adjustment based on signal statistics, removing the burden of manual detection algorithm design while maintaining high detection accuracy.
Solution Approach 2:
The invention replaces manual detection algorithm design (mechanical process) with automated machine learning (intelligent process). Instead of engineers manually designing detection algorithms for each scenario, the system uses unsupervised learning to automatically adapt, substituting human effort with automated intelligent processing.
3Measurement precision
If manual algorithm design with pre-defined characteristics is used, then detection performance for known scenarios is improved, but ease of operation and deployment deteriorates
Solution Approach 1:
The system automatically adapts to different operational scenarios without requiring manual configuration or expert intervention. The unsupervised learning algorithm continuously learns from the received signals and adjusts its detection parameters autonomously, making the system easy to deploy and operate while maintaining high detection performance across various jammer types.
4Reliability
If fixed detection algorithms are used, then reliability for anticipated scenarios is improved, but adaptability to changing jammer characteristics deteriorates
Solution Approach 1:
The system transitions from static fixed detection algorithms to dynamic adaptive algorithms. The unsupervised learning module continuously adjusts its detection parameters based on the statistical properties of the received signals, enabling the system to maintain high reliability while adapting to changing jammer characteristics in real-time.
Solution Approach 2:
The invention enables dynamic parameter changes in the detection algorithm by using unsupervised learning. The system automatically adjusts its detection parameters based on the learned signal characteristics, allowing it to maintain reliability across different scenarios without requiring fixed pre-defined characteristics.
Data Source
Figure 1
Figure 2~4
Figure 5~6
AI summary
The invention relates to a method of training a machine learning module (10) for detecting at least one jammed frequency slot in a frequency hopping signal, wherein the method comprises the steps of: - Generating IQ samples associated with a jammed frequency hopping baseband signal having hops and slots, wherein a pre-defined number of the hops and slots is jammed, - Labeling the IQ samples generated, thereby obtaining labels indicating at least one of jammed hops and/or benign hops and jammed slots and/or benign slots, and - Training the machine learning module (10) by using the IQ samples generated and the labels obtained, wherein the machine learning module (10) is configured to execute an artificial neural network that is trained. Further, the invention relates to a receiver.