Radar emitter multi-label recognition based on residual network

In low signal-to-noise ratio(SNR)environments,the traditional radar emitter recognition(RER)method struggles to recognize multiple radar emitter signals in parallel.This paper proposes a multi-label classification and recognition method for multiple radar-emitter modulation types based on a residual network.This method can quickly perform parallel classification and recognition of multi-modulation radar time-domain aliasing signals under low SNRs.First,we perform time-frequency analysis on the received signal to extract the normalized time-frequency image through the short-time Fourier transform(STFT).The time-frequency distribution image is then denoised using a deep normalized convolutional neural network(DNCNN).Secondly,the multi-label classification and recognition model for multi-modulation radar emitter time-domain aliasing signals is established,and learning the characteristics of radar signal time-frequency distribution image dataset to achieve the purpose of training model.Finally,time-frequency image is recognized and classified through the model,thus completing the automatic classification and recognition of the time-domain aliasing signal.Simulation results show that the proposed method can classify and recognize radar emitter signals of different modulation types in parallel under low SNRs.

Radar Emitter Signal Recognition Based on Complexity Features

Intra-pulse characteristics of different radar emitter signals reflect on signal waveform by way of changing frequency, phase and amplitude. A novel approach was proposed to extract complexity features of radar emitter signals in a wide range of signal-to-noise ratio (SNR), and radial basis probability neural network (RBPNN) was used to recognize different radar emitter signals. Complexity features, including Lempel-Ziv complexity (LZC) and correlation dimension (CD), can measure the complexity and irregularity of signals, which mirrors the intra-pulse modulation laws of radar emitter signals. In an experiment, LZC and CD features of 10 typical radar emitter signals were extracted and RBPNN was applied to identify the 10 radar emitter signals. Simulation results show that the proposed approach is effective and has good application values because average accurate recognition rate is high when SNR varies in a wide range.

Radar Emitter Signal Recognition Using Wavelet Packet Transform and Support Vector Machines

This paper presents a novel method for radar emitter signal recognition. First, wavelet packet transform （WPT） is introduced to extract features from radar emitter signals. Then, rough set theory is used to select the optimal feature subset with good discriminability from original feature set, and support vector machines （SVMs） are employed to design classifiers. A large number of experimental results show that the proposed method achieves very high recognition rates for 9 radar emitter signals in a wide range of signal-to-noise rates, and proves a feasible and valid method.

A three-way incremental-learning algorithm for radar emitter identification

Radar emitter identification has been recognized as an indispensable task for electronic intelligence system. With the increasingly accumulated radar emitter intelligence and information, one key issue is to rebuild the radar emitter classifier efficiently with the newly-arrived information. Although existing incremental learning algorithms are superior in saving significant computational cost by incremental learning on continuously increasing training samples, they are not adaptable enough yet when emitter types , features and sampies are increasing dramatically. For instance, the intra-pulse characters of emitter signals could be fitrther extracted and thus expand the feature dimension. The; same goes for the radar emitter type dimension when samples from new radar emitter types are gathered. In addition, existing incremental classifiers are still problematic in terms of computational cost, sensitivity to data input order, and difficulty in multiemitter type identification. To address the above problems, we bring forward a three-way incremental learning algorithm （TILA） for radar emitter identification which is adaptable for the increase in emitter features, types and. samples.

Multi-function radar emitter identification based on stochastic syntax-directed translation schema

To cope with the problem of emitter identification caused by the radar words＇ uncertainty of measured multi-function radar emitters, this paper proposes a new identification method based on stochastic syntax-directed translation schema（SSDTS）. This method, which is deduced from the syntactic modeling of multi-function radars, considers the probabilities of radar phrases appearance in different radar modes as well as the probabilities of radar word errors occurrence in different radar phrases. It concludes that the proposed method can not only correct the defective radar words by using the stochastic translation schema, but also identify the real radar phrases and working modes of measured emitters concurrently. Furthermore, a number of simulations are presented to demonstrate the identification capability and adaptability of the SSDTS algorithm.The results show that even under the condition of the defective radar words distorted by noise,the proposed algorithm can infer the phrases, work modes and types of measured emitters correctly.

NOVEL RADAR SIGNAL SORTING METHOD BASED ON GEOMETRIC COVERING

With the increase of complexity of electromagnetic environment and continuous appearance of advanced system radars,signals received by radar reconnaissance receivers become even more intensive and complex.Therefore,traditional radar sorting methods based on neural network algorithms and support vector machine(SVM) cannot process them effectively.Aiming at solving this problem,a novel radar signal sorting method based on the cloud model theory and the geometric covering algorithm is proposed.By applying the geometric covering algorithm to divide input signals into different covering domains based on their distribution characteristics,the method can overcome a typical problem that it is easy for traditional sorting algorithms to fall into the local extrema due to the use of complex nonlinear equation to describe input signals.The method uses the cloud model to describe the membership degree between signals to be sorted and their covering domains,thus it avoids the disadvantage that traditional sorting methods based on hard clustering cannot deinterleave the signal samples with overlapped parameters. Experimental results show that the presented method can effectively sort advanced system radar signals with overlapped parameters in complex electromagnetic environment.