Most modern radar systems use a monostatic configuration and exploit wide bandwidth to achieve high down range resolution used for target detection and classification. However increasing pressure on occupancy of the radio spectrum and the requirement for ever more accurate target classification poses serious challenges to future radar designs. Techniques based on narrow band radar are thus being investigated as a mean of reducing spectral occupancy. This approach can be coupled with the use of a multiple radar (multistatic) geometry to provide a potentially powerful technique for improving target detection and classification even beyond that of conventional systems. A multistatic topography allows the application of tomographic techniques for target imaging. Tomography is a process by which a two-dimensional cross-sectional image of an object is obtained via illumination from a variety of differing angles in a variety of differing planes. In radar tomography observations from multiple radar locations enable a three dimensional projection in Fourier space. In this way a three dimensional image of an object can be constructed using techniques such as Backprojection. The use of a narrow band waveform in multistatic radar tomography trades resolution achieved by bandwidth for resolution achieved by spatially diverse angular imaging. This thesis reports a detailed investigation into a range of narrow band, multistatic geometries and techniques to obtain high resolution imaging of moving targets. Images processed using Synthetic Aperture Radar (SAR) and Inverse SAR (ISAR) configurations, modified to emulate multistatic narrow band configurations, have been investigated for both real and simulated data. The effect on the spatial resolution due to masking, ambiguity and coherency of targets consisting of both sparse and dense scatterers was analysed under a range of conditions. A cross range resolution of A/4 was achieved using simulated data. This analysis was also extended to the case of real data of typical ground targets. In this situation the data is inevitably significantly affected by noise but a resolution of A/2 was achieved. This study concludes with a comparison of modeled narrow band system performance with theoretical predictions leading to a preliminary assessment of the capability of the narrow band radar tomographic imaging technique for potential applications.

Narrow band high resolution radar imaging.

Fault diagnosis plays a very important role in ensuring the safe and reliable operations of machines. Currently, the deep learning-based fault diagnosis is attracting increasing attention. However, fault diagnosis under variable working conditions has been a significant challenge due to the domain discrepancy problem. This problem is also unavoidable in deep learning-based fault diagnosis methods. This paper contributes to the ongoing investigation by proposing a new approach for the fault diagnosis under variable working conditions based on STFT and transfer deep residual network (TDRN). The STFT was employed to convert vibration signal to time-frequency image as the input of the TDRN. To address the domain discrepancy problem, the TDRN was developed in this paper. Unlike traditional deep convolutional neural network (DCNN) methods, by combining with transfer learning, the TDRN can make a bridge between two different working conditions, thereby using the knowledge learned from a working condition to achieve a high classification accuracy in another working condition. Moreover, since the residual learning is introducing, the TDRN can overcome the problems of training difficulty and performance degradation existing in traditional DCNN methods, thus further improving the classification accuracy. Experiments are conducted on the popular CWRU bearing dataset to validate the effectiveness and superiority of the proposed approach. The results show that the developed TDRN outperforms those methods without transfer learning and/or residual learning in terms of the accuracy and feature learning ability for the fault diagnosis under variable working conditions.

/pdf/fault-diagnosis-under-variable-working-conditions-based-on-30x0psy3af.pdf

Fault Diagnosis under Variable Working Conditions Based on STFT and Transfer Deep Residual Network

Seismic signal sparse time–frequency representation by Lp-quasinorm constraint

Rolling bearings are important components of rotating machines For their preventive maintenance, it is not enough to know whether there is any fault or the fault type For an effective maintenance, a fault severity monitoring needs to be conducted Currently, the bearing fault diagnosis method based on time–frequency image (TFI) recognition is attracting increasing attention This paper contributes to the ongoing investigation by proposing a new approach for the fault severity monitoring of rolling bearings based on the texture feature extraction of sparse TFIs The first and main step is to obtain accurate TFIs from the vibration signals of rolling bearings Traditional time–frequency analysis methods have disadvantages such as low resolution and cross-term interference Therefore, the TFIs obtained cannot satisfactorily express the time–frequency characteristics of bearing vibration signals To solve this problem, a sparse time–frequency analysis method based on the first-order primal-dual algorithm (STFA-PD) was developed in this paper Unlike traditional time–frequency analysis methods, the time–frequency analysis model of the STFA-PD method is based on the theory of sparse representation, and is solved using the first-order primal-dual algorithm For employing the sparse constraint in the frequency domain, the STFA-PD obtains a higher time–frequency resolution and is free from cross-term interference, as the model is based on a linear time–frequency analysis method The gray level co-occurrence matrix is then employed to extract texture features from the sparse TFIs as input features for classifiers Vibration signals of rolling bearings with different fault severity degrees are used to validate the proposed approach The experimental results show that the developed STFA-PD outperforms traditional time–frequency analysis methods in terms of the accuracy and effectiveness for the fault severity monitoring of rolling bearings

Fault Severity Monitoring of Rolling Bearings Based on Texture Feature Extraction of Sparse Time–Frequency Images

This paper focuses on developing an anti-velocity jamming strategy that enhances the ability of a pulse-Doppler (PD) radar to detect moving targets in the presence of translational and/or micro motion velocity jamming generated by the digital radio frequency memory (DRFM) repeat jammers. The strategy adopts random pulse initial phase (RPIP) pulses as its transmitted signal and thus gets DRFM jammers not adaptable to the randomness of initial phase of the transmitted pulses in the pulse repetition interval (PRI) domain. The difference between the true target echo and the false target jamming signal at each PRI is then utilized to recognize the true and false target signals. In particular, an entropy based multi-channel processing scheme is designed to extract the information of the received signal without the assumption that true and false targets must be both included within one coherent processing interval (CPI). Information such as the component of the received signal (target echo only, jamming only or both) or the operating manner of DRFM repeat jammer can be gained (if jamming exists). Meanwhile, we solve the false target recognition problem under sparse theory frame and our previous work named the short-time sparse recovery (STSR) algorithm is introduced to recover the motion parameters of the true and/or false targets in the time-frequency domain. It should be pointed out that both the translational false target jamming and micro motion target jamming can be recognized in our strategy. The performance of the proposed strategy is compared with the correlated processing (CP) method used by most extant strategies. It is shown that the proposed strategy can successfully recognize the existence of true and/or false targets and keep its power in recovering corresponding motion parameters even when the jamming environment is strong.

https://www.mdpi.com/1424-8220/18/4/1249/pdf

A Sparse-Driven Anti-Velocity Deception Jamming Strategy Based on Pulse-Doppler Radar with Random Pulse Initial Phases.

This paper focuses on the high resolution time-frequency distribution (TFD) of multicomponent signals with amplitude and frequency modulations, and a concise method named short-time sparse representation (STSR) is proposed. In STSR, both analysis and synthesis of the discrete signal can be achieved by exploiting the signal's sparsity in frequency domain at each time instant. In order to fasten the STSR procedure, an efficient sparse recovery algorithm named SL0 is applied, and the signal model for each sliding window is modified to form the same dictionary, which guarantees that the whole recovery procedure adapts to the matrix form. The performance of STSR is compared with other TFD techniques and assessed in various configurations. It is shown that both preferable representation and acceptable computational cost can be obtained.

High Resolution Time-Frequency Distribution Based on Short-Time Sparse Representation

The effect of platform skin return on multiple-element retrodirective cross-eye jamming (MRCJ) system is investigated, allowing the differences between multiple jammer loops to be considered. The jammer antennas employ a linear array with a common center. The median value of the total cross-eye gain and the jamming-to-signal ratio (JSR) are derived to explore their relationship. Simulation results demonstrate the performance of MRCJ in the presence of platform skin return.

Platform skin return and multiple-element linear retrodirective cross-eye jamming

The random pulse initial phase radar has potential applications in suppressing digital radio frequency memory (DRFM) repeat jamming, due to the significant differences between the repeat jamming and the true target echo in phases. However, owing to phase randomness and existence of false targets, the high sidelobe pedestal generated in correlation processing will decrease the performance of target detecting. Considering the low sidelobe characteristic under the framework of compressed sensing (CS), a novel sparsity-driven strategy for false target identification is proposed in this paper. Since the differences mentioned above will induce that the echoes of true and false targets are sparse in different dictionaries, the radar echo can be sparse represented in the union of two kinds of dictionaries, and the true and false targets can then be separated. According to the simulation results, we can see that the proposed method shows high performance on detecting targets and high accuracy in evaluating the motion parameters of both the true and false targets.

Velocity false target identification in random pulse initial phase radar based on compressed sensing

The invention provides a long frequency band radar target jogging feature estimation method. The technical scheme is that a target narrow band echo obtained by a low frequency band radar is subjected to N point discrete sampling to obtain discrete signals, then complex logarithm operation is performed on the discrete signals, phase modulation signals are obtained by phase ambiguity resolving and are subjected to frequency spectrum estimation, and a target jogging spectrum is obtained by calculating, so the jogging feature of the target is obtained. The long frequency band radar target jogging feature estimation method is suitable for a low frequency radar small-amplitude jogging target, the estimation precision of jogging parameters is high, applicable systems have wide frequency bands and the signal to noise ratio demand is low.

Long frequency band radar target jogging feature estimation method

Few-Shot Image Classification (FSIC) aims to learn an image classifier with only a few training samples. The key challenge of few-shot image classification is to learn this classifier with scarce labeled data. To tackle the issue, we leverage the self-supervised learning (SSL) paradigm to exploit unsupervised information. This work builds upon two-stage training paradigm, to push the current state-of-the-art (SOTA) in solving FSIC problem further. Specifically, we incorporate the traditional self-supervised learning method (TSSL) into the pre-training stage and propose an episodic contrastive loss (CL) as an auxiliary supervision for the meta-training stage. The proposed bipartite method, called FSIC-SSL, can SOTA task accuracies on two mainstream FSIC benchmark datasets. Our code will be available at https://github.com/SethDeng/FSIC_SSL . 

Dongping Liao

Papers

High Resolution Time-Frequency Distribution Based on Short-Time Sparse Representation

Platform skin return and multiple-element linear retrodirective cross-eye jamming

Velocity false target identification in random pulse initial phase radar based on compressed sensing

Long frequency band radar target jogging feature estimation method

Improving Few-Shot Image Classification with Self-supervised Learning