Interference Signal Feature Extraction and Pattern Classification Algorithm Based on Deep Learning

Abstract

Aiming at the scarcity of Low Earth Orbit (LEO) satellite spectrum resources, this paper proposes an algorithm of interference signal feature extraction and pattern classification based on deep learning to further improve the stability of satellite–ground communication links. The algorithm can successfully predict the interference signal pattern, start–stop time, frequency change range and other parameters, and has the advantages of excellent interference detection performance, high detection accuracy and small parameter prediction error, etc. It can be applied in the field of channel monitoring of communication satellite-to-ground communication links, and realize the repeated and efficient utilization of spectrum resources. Experiments show that the precision and recall of the algorithm for detecting five kinds of interference signals are all close to 100%, the prediction error of starting and ending time is less than 4 ms, and the prediction error of starting and ending frequency is less than 6 KHz.

On Integrated Cooperative Radio Sensing for Spatial Electromagnetic Analysis in 6G

“…RF signal classification can be seen as a kind of spectrum sensing which provides not only information about the occupancy of a specific frequency range but also about the type of the signal. Hence, with the help of signal classification, interference management and differentiation between friendly and hostile radios can be significantly improved [24]. If the decision about a channel's status depends only on the detected energy, jammers can easily block the resources through constant transmission.…”

GNSS blanket jamming classification algorithm based on spatial attention mechanism and residual shrinkage neural network

“…Mainstream approaches to jamming signal classification can be divided into two categories: machine learning-based approaches and deep learning-based approaches. Typically, machine learning-based jamming classification techniques involve manual extraction of signal features such as kurtosis, root mean square, energy, and entropy by means of varied statistical and mathematical techniques, whilst also training classifiers such as support vector machines, BP neural networks, or decision trees [4][5][6][7][8]. However, the classification accuracy of these schemes is low when the signal jamming patterns are extremely complex and diverse, due to the inevitable loss of some important information in the feature extraction process.…”