Decentralized Automatic Modulation Classification Method Based on Lightweight Neural Network

IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC), pp. 259–264

Abstract

Due to the computing capability and memory limitations, it is difficult to apply the traditional deep learning (DL) models to the edge devices (EDs) for realizing automatic modulation classification (AMC). In this paper, a lightweight neural network for decentralized learning-based automatic modulation classification (DecentAMC) method is proposed. Specifically, group convolutional neural network (GCNN) is designed by replacing the standard convolution layer with the group convolution layer, replacing the flatten layer with the global average pooling (GAP) layer and removing part of fully connected layers. DecentAMC method is achieved by the cooperation in which multiple EDs update and upload the model weight to a central device (CD) for model aggregation to avoid the data privacy disclosure. Experimental results show that the proposed GCNN-based DecentAMC method can improve training efficiency to about 4 times and 57 times than that of GCNN-based centralized AMC (CentAMC) and CNN-based DecentAMC respectively. GCNN-based DecentAMC method can effectively reduce the communication cost and save storage of EDs when compared with CNN-based DecentAMC. Meanwhile, the time complexity and the space complexity of GCNN is significantly decreased when compared with CNN and SCNN, which is suitable to be deployed in EDs.

Authors 7

1. Biao Dong

   Nanjing University of Posts and Telecommunications

   Affiliation as printed

   College of Telecommunications and Information Engineering, NJUPT,Nanjing,China

   College of Telecommunications and Information Engineering, NJUPT, Nanjing, China

2. Guozhen Xu

   National University of Defense Technology

   Affiliation as printed

   College of Electronic countermeasure, National University of Defense Technology,Hefei,China

   College of Electronic countermeasure, National University of Defense Technology, Hefei, China

3. Xue Gang Fu

   Nanjing University of Posts and Telecommunications

   Affiliation as printed

   College of Telecommunications and Information Engineering, NJUPT,Nanjing,China

   College of Telecommunications and Information Engineering, NJUPT, Nanjing, China

4. Heng Dong

   Nanjing University of Posts and Telecommunications

   Affiliation as printed

   College of Telecommunications and Information Engineering, NJUPT,Nanjing,China

   College of Telecommunications and Information Engineering, NJUPT, Nanjing, China

5. Guan Gui

   Nanjing University of Posts and Telecommunications

   Affiliation as printed

   College of Telecommunications and Information Engineering, NJUPT,Nanjing,China

   College of Telecommunications and Information Engineering, NJUPT, Nanjing, China

6. Haris Gacanin Aachen Faculty of Electrical Engineering and Information Technology

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University,Faculty of Electrical Engineering and Information Technology,Aachen,Germany

   Faculty of Electrical Engineering and Information Technology, RWTH Aachen University, Aachen, Germany

7. Fumiyuki Adachi

   Tohoku University

   Affiliation as printed

   Tohoku University,Research Organization of Electrical Communication,Sendai,Japan

   Research Organization of Electrical Communication, Tohoku University, Sendai, Japan