Convolutional Neural Networks for Automatic Modulation Classification

Kaya O., Güçlüoğlu T.

Avrupa Bilim ve Teknoloji Dergisi, vol.44, no.1, pp.93-96, 2022 (Conference Book)

  • Publication Type: Article / Article
  • Volume: 44 Issue: 1
  • Publication Date: 2022
  • Doi Number: 10.31590/ejosat.1224925
  • Journal Name: Avrupa Bilim ve Teknoloji Dergisi
  • Journal Indexes: Index Copernicus, Sobiad Atıf Dizini
  • Page Numbers: pp.93-96
  • Yıldız Technical University Affiliated: Yes

Abstract

Automatic Modulation Classification (AMC) is the process of determining the modulation type of the signal which is taken by receiver in a communication system. Deep learning is a machine learning method having recently attracted great attention due to its superior performance of classifying complex data structures. The automatic modulation classification process, having a critical role in both civil and military applications, is examined in this study by using Convolutional Neural Networks (CNN), one of the deep learning approaches. Within this scope, the performance of the changes made on the net has been interpreted for different signal-to-noise ratio (SNR) values.