TY - GEN
T1 - Lightweight network and model aggregation for automatic modulation classification in wireless communications
AU - Fu, Xue
AU - Gui, Guan
AU - Wang, Yu
AU - Ohtsuki, Tomoaki
AU - Adebisi, Bamidele
AU - Gacanin, Haris
AU - Adachi, Fumiyuki
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - This paper proposes a decentralized automatic modulation classification (DecentAMC) method using light network and model aggregation. Specifically, the lightweight network is designed by separable convolution neural network (S-CNN), in which the separable convolution layer is utilized to replace the standard convolution layer and most of the fully connected layers are cut off, the model aggregation is realized by a central device (CD) for edge device (ED) model weights aggregation and multiple EDs for ED model training. Simulation results show that the model complexity of S-CNN is decreased by about 94% while the average CCP is degraded by less than 1% when compared with CNN and that the proposed AMC method improves the training efficiency when compared with the centralized AMC (CentAMC) using S-CNN.
AB - This paper proposes a decentralized automatic modulation classification (DecentAMC) method using light network and model aggregation. Specifically, the lightweight network is designed by separable convolution neural network (S-CNN), in which the separable convolution layer is utilized to replace the standard convolution layer and most of the fully connected layers are cut off, the model aggregation is realized by a central device (CD) for edge device (ED) model weights aggregation and multiple EDs for ED model training. Simulation results show that the model complexity of S-CNN is decreased by about 94% while the average CCP is degraded by less than 1% when compared with CNN and that the proposed AMC method improves the training efficiency when compared with the centralized AMC (CentAMC) using S-CNN.
KW - Automatic modulation classification (AMC)
KW - Lightweight network
KW - Model aggregation
UR - http://www.scopus.com/inward/record.url?scp=85119377811&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85119377811&partnerID=8YFLogxK
U2 - 10.1109/WCNC49053.2021.9417592
DO - 10.1109/WCNC49053.2021.9417592
M3 - Conference contribution
AN - SCOPUS:85119377811
T3 - IEEE Wireless Communications and Networking Conference, WCNC
BT - 2021 IEEE Wireless Communications and Networking Conference, WCNC 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2021 IEEE Wireless Communications and Networking Conference, WCNC 2021
Y2 - 29 March 2021 through 1 April 2021
ER -