Modulation Recognition using Wavelet-assisted Convolutional Neural Network

Zhang, Qianwen; Xu, Zhan; Zhang, Peiyue

doi:10.1109/atc.2018.8587447

Cited by 10 publications

(1 citation statement)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Potential applications of the proposed method include health monitoring [46,47], machine-human interface [48], activity recognition [45], machinery fault diagnosis [19,17,18], audio signal processing [27,20,10], radar signal processing [36], automatic modulation recognition [49,50], wireless sensor networks and internet of things [51,52], and communication systems [38,37], to name a few. In these applications, data is collected as a time series and fed to an ML model to perform a specific task such as classification, regression, or forecasting.…”

Section: Related Workmentioning

confidence: 99%

Design and Analysis of Convolutional Neural Layers: A Signal Processing Perspective

Farag

2023

IEEE Access

View full text Add to dashboard Cite

Convolutional layers (CLs) are ubiquitous in contemporary deep neural network (DNN) models, commonly used for automatic feature extraction. A CL performs cross-correlation between the input to the layer and a set of learnable kernels to produce the layer output. Typically, kernel weights are randomly initialized and automatically learned during model training using the backpropagation and gradient descent algorithms to minimize a specific loss function. Modern DNN models comprise deep hierarchical stacks of CLs and pooling layers. Despite their prevalence, CLs are perceived as a magical tool for feature extraction without solid interpretations of their underlying working principles. In this work, we advance a method for designing and analyzing CLs by providing novel signal processing interpretations of the CL by exploiting the correlation and equivalent convolution functions of the layer. The proposed interpretations enable the employment of CLs to develop finite impulse response (FIR) filters, matched filters (MFs), short-time Fourier transform (STFT), discrete-time Fourier transform (DTFT), and continuous wavelet transform (CWT) algorithms. The main idea is to pre-assign the CL kernel weights to implement a specific convolution-or correlation-based DSP algorithm. Such an approach enables building self-contained DNN models in which CLs are utilized for various preprocessing and feature extractions tasks, enhancing the model portability, and cutting down the preprocessing computational cost. The proposed DSP interpretations provide an effective means to analyze and explain the operation of automatically trained CLs in the time and frequency domains by reversing the design procedures. The presented interpretations are mathematically established and experimentally validated with a comprehensive machinery fault diagnosis application example illustrating the potential of the proposed methodology.

show abstract

Section: Related Workmentioning

confidence: 99%

Design and Analysis of Convolutional Neural Layers: A Signal Processing Perspective

Farag

2023

IEEE Access

View full text Add to dashboard Cite

show abstract

Automatic Modulation Classification in Time-Varying Channels Based on Deep Learning

2020

View full text Add to dashboard Cite

Automatic modulation classification (AMC) is an important technology in military signal reconnaissance and civilian communications such as cognitive radios. Most of the existing works focused on the AMC in additional white Gaussian noise channels, but the AMC in time-varying wireless channels is more practical and challenging. In this paper, we investigate the AMC in time-varying channels by using the deep learning method for high classification accuracy. Specifically, we take the modulation constellation diagram (CD) as the key feature and propose a slotted constellation diagram (slotted-CD) scheme in order to extract the feature of the time-evolution of the CD due to channel variation. We then develop an advanced neural network for modulation classification, where the output sub-images from the slotted-CD feature extractor are first processed separately by a number of parallel convolutional neural networks and then further processed by a recurrent neural network for exploring their time relationship. Experimental results show that the proposed AMC scheme achieves higher classification accuracy in both slow and fast fading channels when compared with the traditional deep learning based AMC schemes. Such performance improvement can be clearly illustrated by visualizing the outputs of the convolutional layers of the classifier. We also show that visualization can help optimize the parameters of the AMC neural networks.INDEX TERMS Automatic modulation classification, Constellation diagram, Time-varying, Convolutional neural network, Bidirectional long short-term memory network.

show abstract