Few-Shot Modulation Classification Method Based on Feature Dimension Reduction and Pseudo-Label Training

Shi, Yunhao; Xu, Hua; Jiang, Lei; Liu, Yinghui

doi:10.1109/access.2020.3012712

Cited by 10 publications

(6 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More importantly, the raw IQ input was not considered in their hybrid structure which may lead to severe performance loss. The authors in [51] focused on the semisupervised learning scenario, where some handcrafted features, such as high-order cumulants features, entropy features and time-frequency features were combined with unsupervised features extracted by autoencoder as well as the labeled samples to train an annotator to label the unlabeled samples. As a result, adequate pseudo-labeled samples and a few real-labeled samples were applied to train a classifier.…”

Section: Hybrid Amc Methodsmentioning

confidence: 99%

Toward Next-Generation Signal Intelligence: A Hybrid Knowledge and Data-Driven Deep Learning Framework for Radio Signal Classification

Zheng¹,

Zhou

Zhang³

et al. 2023

IEEE Trans. Cogn. Commun. Netw.

View full text Add to dashboard Cite

Automatic modulation classification (AMC) can generally be divided into knowledge-based methods and data-driven methods. In this paper, we explore combining the knowledgebased method and data-driven technology to take full advantage of both and propose a hybrid knowledge and data-driven deep learning framework (HKDD) for AMC. To make the handcrafted features more discriminative, various traditional features are adopted, including instantaneous features, statistical features, and spectral features. In the HKDD framework, a feature fusion mechanism is proposed to integrate the features learned from the original signal with those processed by a fully connected network from the handcrafted features. Besides, an attention mechanism is implemented on the fused features to neglect immature features and highlight important features. To evaluate the performance of the proposed method, we construct two modulation classification datasets containing both traditional features and raw IQ data. The bigger one contains 36 modulation categories, which is greater than the number of categories of any AMC dataset currently available. Simulation results show that our proposed method has significant performance gain in both adequate-sample classification scenario and few-shot classification scenario.

show abstract

Section: Hybrid Amc Methodsmentioning

confidence: 99%

Toward Next-Generation Signal Intelligence: A Hybrid Knowledge and Data-Driven Deep Learning Framework for Radio Signal Classification

Zheng¹,

Zhou

Zhang³

et al. 2023

IEEE Trans. Cogn. Commun. Netw.

View full text Add to dashboard Cite

show abstract

“…Different deep RESNET models differ in convolution layers including ResNet-18, ResNet-34, ResNet-50, ResNet-101 and ResNet-152. RESNET can be used to train the deep convolution neural network, and the network framework is relatively simple [9].…”

Section: Proposed Methodsmentioning

confidence: 99%

Detection of pig based on improved RESNET model in natural scene

Song

Fang

Wang

et al. 2021

Applied Mathematics and Nonlinear Sciences

View full text Add to dashboard Cite

The behaviours of the pig are often closely related to their health. Pig recognition is very important for pig behaviour analysis and digital breeding. Currently, the early signs and abnormal behaviours of sick pigs in breeding farms are mainly completed by human observation. However, visual inspection is labour intensive and time-consuming, and it suffers from the problems of individual experiences and varying environments. An improved ResNet model was proposed and applied to detect individual pigs in this study based on deep learning knowledge. The developed model captured the features of pigs applying across layer connections, and the ability of feature expression was improved by adding a new residual module. The number of layers was reduced to minimise the net complexity. Generally, the ResNet frame was developed by reducing the number of convolution layers, constructing different types of the residual module and adding the number of convolution kernels. The training accuracy and testing accuracy reached 98.2% and 96.4%, respectively, when using the improved model. The experiment results showed that the method proposed in this paper for checking living situations and disease prevention of commercial pigs in pig farms is potential.

show abstract

“…In the cases where unlabeled data are available, pseudolabeling can be adopted to label such data, as shown in [50]. Before pseudo-labeling, the feature set consisting of 10 handcrafted features and 30 AutoEncoder (AE)-learned features is optimized to remove redundant and irrelevant features by using a fast correlation-based filter.…”

Section: E Data Augmentationmentioning

confidence: 99%

“…The optimal feature set is fed to a pseudo-label annotator, which will label an unlabeled sample if its first Softmax probability is higher than the sum of the second and third Softmax probabilities. However, [50] assumes 100 labeled samples are available for each class at each SNR. Moreover, the applied policy for pseudo-labeling cannot guarantee that the selected label is correct, especially when applied to signals with unseen transmitter and channel parameters and hardware imperfections.…”

Section: E Data Augmentationmentioning

confidence: 99%

Channel and Hardware Impairment Data Augmentation for Robust Modulation Classification

Perenda,

Bovet,

Zheleva

et al. 2024

IEEE Trans. Cogn. Commun. Netw.

View full text Add to dashboard Cite

Deep learning has achieved remarkable results in modulation classification under two assumptions: a large amount of labeled class-balanced data is available, and test data shares the same distribution as data used for training. However, due to dynamic wireless fading channels and hardware imperfections, it is implausible that these assumptions hold in practice. This paper proposes Model-based Data Augmentation for Deep learningbased Modulation Classification (MDA-DMC) to build a highquality dataset from a small amount of labeled seed data. We develop two novel augmentation methods to combat channel and hardware impairments, along with two widely used augmentation methods, adding Gaussian noise and rotation. We are the first to investigate the relationship between augmentation methods and impairments due to channel and hardware imperfections. We show that MDA-DMC can successfully combat the hardware imperfections for the same channel model as the channel model of seed data. Furthermore, with a few AWGN seed data, MDA-DMC achieves an accuracy gain of up to 17.20% in fading channels. By adding a few fading seed data, MDA-DMC improves accuracy by an additional 10.78% in fading channels.

show abstract

Few-Shot Modulation Classification Method Based on Feature Dimension Reduction and Pseudo-Label Training

Cited by 10 publications

References 32 publications

Toward Next-Generation Signal Intelligence: A Hybrid Knowledge and Data-Driven Deep Learning Framework for Radio Signal Classification

Toward Next-Generation Signal Intelligence: A Hybrid Knowledge and Data-Driven Deep Learning Framework for Radio Signal Classification

Detection of pig based on improved RESNET model in natural scene

Channel and Hardware Impairment Data Augmentation for Robust Modulation Classification

Contact Info

Product

Resources

About