Modulation classification in the presence of adjacent channel interference using convolutional neural networks. (25th February 2020)

• Record Type: Journal Article
• Authors: Al‐Makhlasawy, Rasha M.; Hefnawy, Alaa A.; Abd Elnaby, Mustafa M.; Abd El‐Samie, Fathi E.
• Other Names: Krishna P. Venkata guestEditor.; Yenduri Sumanth guestEditor.; Ariwa Ezendu guestEditor.
• Is Part Of: International journal of communication systems. Volume 33:Number 13(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-02-25
• Subjects: adjacent channel interference (ACI) -- AlexNet -- convolutional neural networks (CNN) -- deep learning (DL) -- VGGNet; Telecommunication systems -- Periodicals; 621.382
• DOI: 10.1002/dac.4295
• Languages: English
• ISSNs: 1074-5351

Abstract

Summary: This paper investigates a vital issue in wireless communication systems, which is the modulation classification. A proposed framework for modulation classification based on deep learning (DL) is presented in the presence of adjacent channel interference (ACI). This framework begins with the generation of constellation diagrams from the received data. These constellation diagrams are fed to convolutional neural networks (CNNs) for modulation classification. The objective of this process is to eliminate the manual feature extraction from the received data and make feature extraction process as a built‐in step with CNNs. Three types of CNNs are considered in this paper and compared for this objective. These types are AlexNet, VGG‐16, and VGG‐19. The proposed classifier is applied on Rayliegh and Rician fading channels. Abstract : The paper presents an efficient approach for modulation classification in wireless communication systems based on deep learning (DL). We use AlexNet, VGG‐16, and VGG‐19 as classifiers. Five types of modulation are considered and classified at different SNR values to validate the proposed approach in different channel scenarios. Fading effect is also considered with adjacent channel interference (ACI) to work in a practical communication scenario. The simulation results show that VGG‐19 has the best performance compared to other classifiers. We can come to a conclusion that the interpretation of a modulation type as a constellation diagram image and the utilization of the evolving deep learning trend for modulation classification in adaptive modulation systems are very promising trends. In addition, the noise effect on the received signals is less severe in the constellation diagrams, which allows the strong CNN classifiers to extract the modulation type from the obtained constellation diagrams.