Jamming Prediction for Radar Signals Using Machine Learning Methods

Abstract

Jamming is a form of electronic warfare where jammers radiate interfering signals toward an enemy radar, disrupting the receiver. The conventional method for determining an effective jamming technique corresponding to a threat signal is based on the library which stores the appropriate jamming method for signal types. However, there is a limit to the use of a library when a threat signal of a new type or a threat signal that has been altered differently from existing types is received. In this paper, we study two methods of predicting the appropriate jamming technique for a received threat signal using deep learning: using a deep neural network on feature values extracted manually from the PDW list and using long short-term memory (LSTM) which takes the PDW list as input. Using training data consisting of pairs of threat signals and corresponding jamming techniques, a deep learning model is trained which outputs jamming techniques for threat signal inputs. Training data are constructed based on the information in the library, but the trained deep learning model is used to predict jamming techniques for received threat signals without using the library. The prediction performance and time complexity of two proposed methods are compared. In particular, the ability to predict jamming techniques for unknown types of radar signals which are not used in the stage of training the model is analyzed.

Artificial Intelligence Aided Electronic Warfare Systems- Recent Trends and Evolving Applications

“…The method can effectively detect and classify the jamming in the low-frequency SAR signals. Two methods of predicting the appropriate jamming technique for a received threat signal using deep learning are presented in [59]. Firstly, a DNN is used on feature values extracted manually from the pulse description width (PDW) list of the radar signal.…”

“…LSTM is mainly used to solve the timing prediction problem because it can predict the state of the next moment based on the state of the data at the previous moment. Hence, it can be used for radar signal processing and for predicting the appropriate jamming technique [59]. 4) Deep reinforcement learning (DRL): Reinforcement learning is an area of machine learning that is used for taking suitable action to maximize the reward in a particular situation.…”

Attribute Classification of Long-Period Pulsed Radar Signals Using CNN-LSTM Networks

“…이러한 레 이다 신호를 분석하기 위해 초기에는 사전에 구축된 신 호 정보의 데이터베이스를 이용했지만 [3] , 최근에 심층학 습을 적용한 방법들이 제시되고 있다. 심층학습은 CNN (convolutional neural network) [4]～ [7] 또는 LSTM(long shortterm memory) 모델 [1], [8] 을 주로 사용하며, 대부분 PRI, RF, PW 등의 범위와 변화 형태와 같은 펄스 레이다 신호의 특징별로 수～수십 가지로 신호를 분류한다. 한편, 참고 문헌 [9]에서는 펄스 레이다 신호를 분석하기 위해 펄스 신호의 주요한 특성인 무선주파수와 펄스 반복 간격의 시간적인 변화 형태를 각각 7가지, 8가지로 구분하고, 변 화 형태별로 속성을 정의한 후, LSTM을 사용하여 변화 형태와 속성을 식별하는 방법을 제시하였다.…”

Machine Learning Algorithms for Identification of Low Signal-to-Noise Ratio Radar Signals

“…The trained classifier from Problem 5 was tested using the simulated single LFM pulse using the parameters given in Table 13. Classification results are shown in Figures 39,40,41,42,43,and 44. Since the parameters are not close to the realistic parameters, the classifiers are not expected to perform perfectly even at high SNR.…”

“…Lee et al [43] investigated jamming prediction for radar signals using a combination of ML classifiers (SVM, NN, and RF) to perform feature extraction and classify the signals. They obtained an accuracy of 98.46% for NN with extracted features, but 99.36% for LSTM without feature extraction.…”