Study On Technology Of Radar Signal Intra-pulse Features Analysis And Recognition

With the rapid development of science and technology,various kinds of new system radars are constantly emerging,and the electromagnetic environment is complex and changeable.How to realize the fast and effective recognition of radar emitter signal in the complex electromagnetic environment has become the key problem of radar electronic countermeasure.In-pulse feature extraction of radar signal is an important means of signal recognition.Deep learning has the advantages of high intelligence and good robustness,which is a research hotspot in the field of radar emitter recognition.Therefore,this article combines deep learning theory and intra-pulse analysis algorithm to study the automatic modulation recognition of modulated signal.The main research work of this paper is as follows:(1)Aiming at the problem that it is difficult to obtain the measured data of radar signals,this paper constructs the simulation data set of radar signals with common modulation types,deduces the complete process of signal-to-noise ratio(SNR)estimation algorithm,introduces the phase unwrapping algorithm and instantaneous frequency estimation algorithm under low SNR,and analyzes the characteristics of instantaneous phase and instantaneous frequency of different modulation signals.At the same time,the image preprocessing and the common classifier model are introduced.(2)In view of the problem that the traditional five parameters have been unable to meet the recognition of radar emitter,this paper proposes a modulation recognition algorithm based on depth transfer learning combined with the instantaneous frequency characteristics of the signal.The average recognition rate reaches 95.75%under the condition of SNR is 0dB,and the radar signal can be recognized effectively.The experimental results show that the proposed method has low computational complexity,and its timeliness and engineering practicability are confirmed.(3)In order to improve the performance of the modulation recognition algorithm,a novel modulation recognition algorithm based on convolutional neural network(CNN)and bilinear pooling(BP)is proposed.A feature fusion method based on BP algorithm.The images of smooth pseudo Wigner-Ville distribution(SPWVD)and Butterworth distribution(BUD)are extracted as the parallel inputs of the neural network.The multilayer convolution block of the residual network is selected as the feature extractor to extract the depth features of the two images.And the BP algorithm is introduced to make full use of the relationship between different time-frequency features to extract complete features from the parallel depth features to complete the recognition.Finally,the proposed method is compared with the traditional classification methods,and the results show that the CNN-BP model can recognize various modulation signals with high accuracy,and has good recognition performance.