Article
Telecommunications
Peihan Qi, Xiaoyu Zhou, Shilian Zheng, Zan Li
Summary: The study proposes a waveform-spectrum multimodal fusion (WSMF) method based on deep residual networks to achieve Automatic Modulation Classification (AMC). By extracting features from multimodal information and using a feature fusion strategy, this method can effectively distinguish between different modulation signals.
IEEE TRANSACTIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING
(2021)
Article
Telecommunications
Xue Fu, Guan Gui, Yu Wang, Tomoaki Ohtsuki, Bamidele Adebisi, Haris Gacanin, Fumiyuki Adachi
Summary: This paper proposes a decentralized learning AMC method using model consolidation and lightweight design, which reduces the storage and computational capacity requirements, improves the training efficiency, and lowers the communication overhead.
IEEE TRANSACTIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING
(2022)
Article
Chemistry, Analytical
Mengtao Wang, Youchen Fan, Shengliang Fang, Tianshu Cui, Donghang Cheng
Summary: This paper proposes a joint automatic modulation discrimination (AMC) model that combines deep learning (DL) and expert features to address the challenges in distinguishing 16QAM and 64QAM. Experimental results show that the proposed model outperforms benchmark networks and improves signal classification accuracy and QAM signal discrimination.
Article
Computer Science, Information Systems
Wei-Tao Zhang, Dan Cui, Shun-Tian Lou
Summary: This paper investigates the application of convolutional neural networks in identifying modulation classes for digitally modulated signals and proposes an effective network structure that outperforms traditional algorithms in terms of classification accuracy.
Article
Computer Science, Information Systems
Shuo Chang, Sai Huang, Ruiyun Zhang, Zhiyong Feng, Liang Liu
Summary: Automatic modulation classification (AMC) plays a vital role in identifying the modulation type of a received signal for ensuring the physical-layer security of IoT networks. This article focuses on reproducing and evaluating popular AMC algorithms using the in-phase/quadrature (I/Q) and amplitude/phase (A/P) representations for comparison. Based on the experimental results, it is found that CNN-RNN-like algorithms using A/P as input data perform better at high signal-to-noise ratio (SNR), while the opposite is true at low SNR. Inspired by these findings, a multitask learning-based deep neural network (MLDNN) is proposed, which effectively fuses I/Q and A/P. Extensive simulations demonstrate the superior performance of the proposed MLDNN in a public benchmark.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Computer Science, Information Systems
Zhongyong Wang, Dongzhe Sun, Kexian Gong, Wei Wang, Peng Sun
Summary: This paper proposes a lightweight convolutional neural network for automatic modulation classification task, with a focus on reducing model complexity by designing depthwise separable convolution residual architecture and using global depthwise convolution for feature reconstruction. Experimental results show significant savings in model parameters and inference time compared to recent works.
Article
Construction & Building Technology
Jiaming Li, Shibin Tang, Kunyao Li, Shichao Zhang, Liexian Tang, Leyu Cao, Fuquan Ji
Summary: The recognition and classification of microseismic waveforms in rock engineering is crucial for predicting instability. Deep learning models like VGG16, ResNet18, AlexNet, and their ensemble model are effective in accurately classifying microseismic waveform images and spectrograms. However, different models have varying performance in recognizing noise, electricity, and microseismic events, requiring careful selection based on real-world scenarios.
TUNNELLING AND UNDERGROUND SPACE TECHNOLOGY
(2022)
Article
Telecommunications
Shanchuan Ying, Sai Huang, Shuo Chang, Zheng Yang, Zhiyong Feng, Ningyan Guo
Summary: In this paper, a data-driven framework called CTDNN is proposed to enhance the classification performance of automatic modulation classification. By using modules such as convolutional neural network, transition module, and final classifier, CTDNN achieves superior classification performance compared to traditional deep models.
CHINA COMMUNICATIONS
(2023)
Article
Chemistry, Analytical
Hui Han, Zhiyuan Ren, Lin Li, Zhigang Zhu
Summary: The paper proposes a feature fusion scheme based on deep learning to address the challenges of background noise and large dynamic input in automatic modulation classification. By fusing features from different domains, the stability and accuracy of signal modulation types are improved.
Article
Computer Science, Artificial Intelligence
Pengcheng Jiang, Yu Xue, Ferrante Neri
Summary: This paper proposes a multi-objective pruning method (MOP-FMS) based on feature map selection, which takes the number of FLOPs as a pruning objective in addition to the accuracy rate. The authors design an efficient search space, domain-specific crossover and mutation operators, decoding and pruning methods, and use multi-objective optimization for evaluation. Experimental results demonstrate that the proposed method achieves higher pruning rate without sacrificing the accuracy rate.
APPLIED SOFT COMPUTING
(2023)
Article
Chemistry, Multidisciplinary
Zhishuang Lin, Qianyu Wang, Chang Lai
Summary: All-sky airglow imagers (ASAIs) are used to study airglow in the middle and upper atmosphere to understand atmospheric perturbation. However, the ripples caused by perturbation are only visible in airglow images taken on clear nights. We trained a convolutional neural network classification model to distinguish between airglow images taken on clear nights and unclear nights, and achieved an accuracy of 99%. The feature maps of five categories also indicate the reliability of the classification model.
APPLIED SCIENCES-BASEL
(2023)
Article
Biology
Assad Rasheed, Arif Iqbal Umar, Syed Hamad Shirazi, Zakir Khan, Shah Nawaz, Muhammad Shahzad
Summary: The healthcare sector is of utmost importance and demands the highest level of service and care. Deep learning has made exciting advancements in the field of medical imaging, particularly in clinical decision support tools. This study introduces a Hybrid model that combines handcrafted features and deep learning techniques to automate the diagnosis of various types of eczema. The results show that deep learning models can accurately classify eczema and perform comparably to dermatologists.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Engineering, Multidisciplinary
Lin Li, Ying Ding, Bo Li, Mengqing Qiao, Biao Ye
Summary: Researchers utilize static and dynamic analysis techniques along with deep learning algorithms to classify malware, achieving high accuracy of 98.68% and a low log loss of 0.022 by integrating features from ASM and section files and fusing them using a double byte feature coding method.
ALEXANDRIA ENGINEERING JOURNAL
(2022)
Article
Computer Science, Information Systems
Hany S. Hussein, Mohamed Hassan Essai Ali, Mohammed Ismeil, Mohamed N. Shaaban, Mona Lotfy Mohamed, Hany A. Atallah
Summary: This study proposes robust CNN-based AMC techniques, which eliminate the need for feature extraction and can achieve high classification accuracy. The developed techniques utilize different classification layers and can automatically learn the features from the transmitted signals during training.
Article
Engineering, Electrical & Electronic
Changbo Hou, Yuqian Li, Xiang Chen, Jing Zhang
Summary: Signal automatic modulation classification plays an important role in military and civilian fields, but still faces many challenges in complex wireless communication environments. To address these challenges, an algorithm based on feature fusion of convolutional neural network is proposed in this paper, which effectively improves the modulation classification performance.
PHYSICAL COMMUNICATION
(2021)