Automatic Modulation Classification Scheme Based on LSTM With Random Erasing and Attention Mechanism

Chen, Yufan; Shao, Wei; Liu, Jin; Yu, Lian; Zhang, Qian

doi:10.1109/access.2020.3017641

Cited by 45 publications

(25 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Occlusion in MOT manifests itself due to the absence of some features of the target. We propose a simulated occlusion strategy based on random erasure [39] applicable to the training process of the LDAE model and simulate object occlusion on the training data. Random erasure is to randomly select a rectangular area in the image and use random values to erase its pixels.…”

Section: Proposed Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Lightweight and Deep Appearance Embedding for Multiple Object Tracking

Zheng

et al. 2022

IET Computer Vision

View full text Add to dashboard Cite

The main challenge of Multiple Object Tracking (MOT) is that there is great uncertainty in data association when using the tracked predicted values and tracked trajectories. Meanwhile, the MOT is complex and time‐consuming. When equipment resources are limited or real‐time requirement is high, its application is very limited. Therefore, we propose a Lightweight Deep Appearance Embedding (LDAE) to assist the association of trajectories. Firstly, in addition to motion information in data association, we also introduce more discriminative appearance features to participate in the affinity measure to effectively distinguish similar targets. Secondly, according to the idea of feature mapping, we design a lightweight deep appearance embedding module. It can help extract appearance features with less computation. Finally, we propose a simulated occlusion strategy for the training of the LDAE, which helps improve the ability to recognise different targets in dense scenes. The LDAE dramatically reduces the computational cost and improves the accuracy of data association. Extensive experiments are conducted on the MOT datasets (MOT16, MOT17 and MOT20), which prove that the LDAE outperforms several state‐of‐the‐art trackers in the tracking accuracy and anti‐occlusion performance. Furthermore, we apply the LDAE to escalators, which can achieve fast and stable tracking effect.

show abstract

Section: Proposed Methodsmentioning

confidence: 99%

“…Occlusion in MOT manifests itself due to the absence of some features of the target. We propose a simulated occlusion strategy based on random erasure [39]…”

Section: Simulated Occlusion Strategymentioning

confidence: 99%

Lightweight and Deep Appearance Embedding for Multiple Object Tracking

Zheng

et al. 2022

IET Computer Vision

View full text Add to dashboard Cite

show abstract

“…where u i represents the i th word in the sequence of target language; c i represents the background vector of the word i [ 20 ]. Assume that the hidden layer state at the moment j of the encoder is H j , and its corresponding background vector can be calculated by

…”

Section: Algorithm Designmentioning

confidence: 99%

A Comparative Study of Text Genres in English-Chinese Translation Effects Based on Deep Learning LSTM

Zhao

Xiao-yan

2022

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

In recent years, neural network-based English-Chinese translation models have gradually supplanted traditional translation methods. The neural translation model primarily models the entire translation process using the “encoder-attention-decoder” structure. Simultaneously, grammar knowledge is essential for translation, as it aids in the grammatical representation of word sequences and reduces grammatical errors. The focus of this article is on two major studies on attention mechanisms and grammatical knowledge, which will be used to carry out the following two studies. Firstly, in view of the existing neural network structure to build translation model caused by long distance dependent on long-distance information lost in the delivery, leading to problems in terms of the translation effect which is not ideal, put forward a kind of embedded attention long short-term memory (LSTM) network translation model. Secondly, in view of the lack of grammatical prior knowledge in translation models, a method is proposed to integrate grammatical information into translation models as prior knowledge. Finally, the proposed model is simulated on the IWSLT2019 dataset. The results show that the proposed model has a better representation of source language context information than the existing translation model based on the standard LSTM model.

show abstract

“…W ITH the development of Deep Learning (DL), many DL-based models, such as Convolutional Neural Networks (CNN) [1]- [6], Long Short-Term Memory Recurrent Neural Networks (LSTM-RNN) [7]- [9], and models [10], [11] combined with CNN and LSTM-RNN, have been proposed for Automatic Modulation Classification (AMC). Although these deep models have achieved great successes and outperformed traditional classifiers, they also bring some hidden security problems.…”

Section: Introductionmentioning

confidence: 99%

Faking Signals to Fool Deep Neural Networks in AMC via Few Data Points

Yang

He³

et al. 2021

IEEE Access

View full text Add to dashboard Cite

The recent years has witnessed a rapid development of Deep Learning (DL) based Automation Modulation Classification (AMC) methods, which has proved to outperform traditional classification approaches. In order to disturb the deep neural networks for AMC, in this paper, we propose an adversarial attack method to generate fake signals for fooling DL-based classifiers. Firstly, some constraints on visual difference and recoverability of fake signals are defined. Next, a Few Data Point Attacker (FDPA) is proposed to generate fake signals with few perturbed data points via differential evolution algorithm. Some experiments are taken on a public dataset, RML 2016.10a, and the results show that fake signals generated by the FDPA can remarkably reduce the accuracies of three types of DL-based AMC classifiers, a Convolutional Neural Network (CNN) based classifier, a Long Short-Term Memory Recurrent Neural Network (LSTM-RNN) based classifier, and a classifier combined with CNN and LSTM-RNN. The code will be available. INDEX TERMS Automatic modulation classification, fake signals, few data point attacker (FDPA), differential evolution.

show abstract

Automatic Modulation Classification Scheme Based on LSTM With Random Erasing and Attention Mechanism

Cited by 45 publications

References 18 publications

Lightweight and Deep Appearance Embedding for Multiple Object Tracking

Lightweight and Deep Appearance Embedding for Multiple Object Tracking

A Comparative Study of Text Genres in English-Chinese Translation Effects Based on Deep Learning LSTM

Faking Signals to Fool Deep Neural Networks in AMC via Few Data Points

Contact Info

Product

Resources

About