Variational Temporal Deep Generative Model for Radar HRRP Target Recognition

  title={Variational Temporal Deep Generative Model for Radar HRRP Target Recognition},
  author={Dandan Guo and Bo Chen and Wenchao Chen and C. Wang and Hongwei Liu and Mingyuan Zhou},
  journal={IEEE Transactions on Signal Processing},
We develop a recurrent gamma belief network (rGBN) for radar automatic target recognition (RATR) based on high-resolution range profile (HRRP), which characterizes the temporal dependence across the range cells of HRRP. The proposed rGBN adopts a hierarchy of gamma distributions to build its temporal deep generative model. For scalable training and fast out-of-sample prediction, we propose the hybrid of a stochastic-gradient Markov chain Monte Carlo (MCMC) and a recurrent variational inference… 

End-to-End Radar HRRP Target Recognition Based on Integrated Denoising and Recognition Network

A hybrid loss is proposed to integrate theDenoising module and the recognition module, which enables end-to-end training, reduces the information loss during denoising, and boosts the recognition performance.

Hierarchical Sequential Feature Extraction Network for Radar Target Recognition Based on HRRP

A novel hierarchical time sequential feature extraction network is proposed and experiments on an airplane electromagnetic calculation dataset were conducted and experimental results showed the superiority of the proposed method.

Deep Variational Graph Convolutional Recurrent Network for Multivariate Time Series Anomaly Detection

This paper model channel dependency and stochasticity within MTS by developing an embedding-guided probabilistic generative network and develops an upward-downward variational inference scheme that considers both forecasting-based and reconstruction-based losses, achieving an accurate posterior ap-proximation of latent variables with better MTS representations.

Switching Gaussian Mixture Variational RNN for Anomaly Detection of Diverse CDN Websites

This work proposes a switching Gaussian mixture variational recurrent neural network (SGmVRNN) suitable for multivariate CDN KPIs and develops an upward-downward autoencoding inference method which combines the bottom-up likelihood and up-bottom prior information of the parameters for accurate posterior approximation.

A Practical Deceptive Jamming Method Based on Vulnerable Location Awareness Adversarial Attack for Radar HRRP Target Recognition

A practical deceptive jamming generation method that learns the vulnerable range cells in an HRRP sample and injects several jamming pulses with specific amplitudes into these range cells to fool the one-dimensional convolutional neural network-based HRRPRATR models.

Spatio-Temporal-Frequency Graph Attention Convolutional Network for Aircraft Recognition Based on Heterogeneous Radar Network

A knowledge-and-data-driven graph neural network-based collaboration learning model for reliable aircraft recognition in a heterogeneous radar network and shows that the expansion of the information dimension can gain considerable benefits to perform robustly in the low signal-to-noise ratio region.

Few-Shot HRRP Target Recognition Based on Gramian Angular Field and Model-Agnostic Meta-Learning

A novel HRRP few-shot recognition method named gramian angular field model-agnostic meta-learning(GAF-MAML) is proposed to solve the above problems and achieved improved performance on an airplane electromagnetic calculation dataset compared with traditional deep learning methods.

W-Band Multi-Aspect High Resolution Range Profile Radar Target Classification Using Support Vector Machines

Millimeter-wave radar measurements were taken for two maritime targets instrumented with attitude and heading reference systems in a littoral environment with the aim of developing a multiaspect classifier, which yielded excellent results in their general forms and were simple to implement.



Bayesian Spatiotemporal Multitask Learning for Radar HRRP Target Recognition

A Bayesian dynamic model based on multitask learning (MTL) is developed for radar automatic target recognition (RATR) using high-resolution range profile (HRRP), where the stick-breaking prior and MTL mechanism are employed to infer the number of hidden states in an HMM and learn the target-dependent states collectively for all targets.

Radar HRRP Target Recognition Based on t-SNE Segmentation and Discriminant Deep Belief Network

Compared with the conventional recognition models, the proposed framework not only makes better use of data set inherent structure among HRRP samples for segmentation, but also utilizes high-level features for recognition.

Radar HRRP target recognition with deep networks

Convolutional neural networks for radar HRRP target recognition and rejection

This paper takes advantage of the attractive properties of convolutional neural networks (CNNs) to address HRRP RATR and rejection problem and devise a two-dimensional CNN model for the spectrogram feature.

Target-Aware Recurrent Attentional Network for Radar HRRP Target Recognition

Max-Margin Deep Generative Models

Empirical results demonstrate that max-margin learning can significantly improve the prediction performance of DGMs and meanwhile retain the generative ability; and mmDGMs are competitive to the state-of-the-art fully discriminative networks by employing deep convolutional neural networks (CNNs) as both recognition and generative models.

Deep Temporal Sigmoid Belief Networks for Sequence Modeling

Deep dynamic generative models are developed to learn sequential dependencies in time-series data. The multi-layered model is designed by constructing a hierarchy of temporal sigmoid belief networks

Noise Robust Radar HRRP Target Recognition Based on Multitask Factor Analysis With Small Training Data Size

A factor analysis model based on multitask learning (MTL) is developed to characterize the FFT-magnitude feature of complex high-resolution range profile (HRRP), motivated by the problem of radar

Convolutional Neural Network With Data Augmentation for SAR Target Recognition

The capability of a deep convolutional neural network (CNN) combined with three types of data augmentation operations in SAR target recognition is investigated, showing that it is a practical approach for target recognition in challenging conditions of target translation, random speckle noise, and missing pose.