MCNet: An Efficient CNN Architecture for Robust Automatic Modulation Classification

Huynh‐The, Thien; Hua, Cam-Hao; Pham, Quoc-Viet; Kim, Dong-Seong

doi:10.1109/lcomm.2020.2968030

Cited by 230 publications

(103 citation statements)

References 13 publications

Supporting

Mentioning

103

Contrasting

Order By: Relevance

“…Finally, the dataset gets 1572864 frames, where the dataset is divided such that 80% of the frames (1258291 frames) are used for training and the remaining (314573 frames) are used for testing. The performance of the proposed model is compared with that of conventional models such as ML-XGboost [39], VGG [39], ResNet [39], CNN-AMC [42], and MCNet [43]. The simulation is performed using an i5 2.9 GHz CPU, 32 GB RAM, and NVIDIA GeForce RTX Table 4.…”

Section: Simulation Resultsmentioning

confidence: 99%

“…Here, AM-DSB-SC cannot be distinguished from 128APSK and 128QAM, and 64QAM cannot be distinguished from AM-SSB-SC. The reference confusion matrices at 10 dB SNR can be referred by [43] To compare the proposed model with the baseline model (MCNet) regarding outstandingness. The overall classification accuracy for each model is summarized in Table 5.…”

Section: Simulation Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Lightweight Deep Learning Model for Automatic Modulation Classification in Cognitive Radio Networks

Kim

Doan

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

Automatic modulation classification (AMC) used in cognitive radio networks is an important class of methods apt to utilize spectrum resources efficiently. However, conventional likelihood-based approaches have high computational complexity. Thus, this paper proposes a novel convolutional neural network architecture for AMC. A bottleneck and asymmetric convolution structure are employed in the proposed model, which can reduce the computational complexity. The skip connection technique is used to solve the vanishing gradient problem and improve the classification accuracy. The dataset DeepSig:RadioML, which is composed of 24 modulation classes, is used for the performance analysis. Simulation results show that the classification accuracy performance of the proposed model is outstanding in the signal-to-noise ratio (SNR) range from-4 dB to 20 dB compared with MCNet that is the best model in the conventional models, where the proposed model achieves 5.52% and 5.92% improvement regarding classification accuracy at the SNRs of 0 dB and 10 dB, respectively. In terms of the computational complexity, the proposed model not only saves the trainable parameters by more than 67% but also reduces the prediction time for a signal by more than 54.4% compared with those of MCNet.

show abstract

Section: Simulation Resultsmentioning

confidence: 99%

Section: Simulation Resultsmentioning

confidence: 99%

Lightweight Deep Learning Model for Automatic Modulation Classification in Cognitive Radio Networks

Kim

Doan

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…AI is a thriving technology for many intelligent applications in various fields. Some high-profile examples of AI are autonomous vehicles (e.g., self-driving car and drones) in automotive, medical diagnosis and telehealth in healthcare, cybersecurity systems (e.g., malware and botnet detection), AI banking in finance, image processing and natural language processing in computer vision, and modulation classification in wireless communications [8]. Among many branches of AI, machine learning (ML) and deep learning (DL) are two important approaches.…”

Section: B Artificial Intelligencementioning

confidence: 99%

“…Among many branches of AI, machine learning (ML) and deep learning (DL) are two important approaches. Generally, ML refers to the ability to learn and extract meaningful patterns from the data, and the performance of ML-based algorithms and systems are heavily dependent on the representative features [8]. In the meanwhile, DL is able to solve complex systems by learning from simple representations.…”

Section: B Artificial Intelligencementioning

confidence: 99%

Artificial Intelligence (AI) and Big Data for Coronavirus (COVID-19) Pandemic: A Survey on the State-of-the-Arts

Pham

Nguyen

Huynh‐The

et al. 2020

Preprint

Self Cite

105

View full text Add to dashboard Cite

The very first infected novel coronavirus case (COVID-19) was found in Hubei, China in Dec. 2019. The COVID-19 pandemic has spread over 215 countries and areas in the world, and has significantly affected every aspect of our daily lives. At the time of writing this article, the numbers of infected cases and deaths still increase significantly and have no sign of a well-controlled situation, e.g., as of 14 April 2020, a cumulative total of 1,853,265 (118,854) infected (dead) COVID-19 cases were reported in the world. Motivated by recent advances and applications of artificial intelligence (AI) and big data in various areas, this paper aims at emphasizing their importance in responding to the COVID-19 outbreak and preventing the severe effects of the COVID-19 pandemic. We firstly present an overview of AI and big data, then identify their applications in fighting against COVID-19, next highlight challenges and issues associated with state-of-the-art solutions, and finally come up with recommendations for the communications to effectively control the COVID-19 situation. It is expected that this paper provides researchers and communities with new insights into the ways AI and big data improve the COVID-19 situation, and drives further studies in stopping the COVID-19 outbreak.

show abstract

“…The recent widespread popularity of machine learning and deep neural network (DNN) permits learning-based signal processing techniques [6]- [8] to achieve a powerful capability of learning features out of data samples [9], which shows great potential in wireless communication systems. In [8], a cost-efficient convolutional neural network (CNN)based method for a robust automatic modulation classification (AMC) is proposed. In particular, it achieves over 93% of 24-modulation at 20 dB through an approach that considers the concurrent learning of the spatiotemporal signal correlation via different asymmetric convolution kernels.…”

Section: Introductionmentioning

confidence: 99%

Deep RP-CNN for Burst Signal Detection in Cognitive Radios

Seo

Nam

2020

IEEE Access

View full text Add to dashboard Cite

This paper proposes a convolutional neural network (CNN)-based signal detection scheme using image encoding techniques for burst signals in wireless networks. The conventional signal detection approach based on energy measurement performs poorly when detecting burst signals owing to the short signal length and relatively long sensing duration. To detect the presence of a burst signal, the proposed scheme encodes the received time-series signal into an image that is further fed to a CNN model. For image encoding techniques, recurrence plot algorithms are adopted in the proposed scheme with a CNN. In particular, the proposed scheme achieves the correct detection probability of 99% even in the presence of a short burst signal at SNR= −10dB.

show abstract

MCNet: An Efficient CNN Architecture for Robust Automatic Modulation Classification

Cited by 230 publications

References 13 publications

Lightweight Deep Learning Model for Automatic Modulation Classification in Cognitive Radio Networks

Lightweight Deep Learning Model for Automatic Modulation Classification in Cognitive Radio Networks

Artificial Intelligence (AI) and Big Data for Coronavirus (COVID-19) Pandemic: A Survey on the State-of-the-Arts

Deep RP-CNN for Burst Signal Detection in Cognitive Radios

Contact Info

Product

Resources

About