A Lightweight Deep Learning Model for Automatic Modulation Classification Using Residual Learning and Squeeze–Excitation Blocks

Abstract

Automatic modulation classification (AMC) is a vital process in wireless communication systems that is fundamentally a classification problem. It is employed to automatically determine the type of modulation of a received signal. Deep learning (DL) methods have gained popularity in addressing the problem of modulation classification, as they automatically learn the features without needing technical expertise. However, their efficacy depends on the complexity of the algorithm, which can be characterized by the number of parameters. In this research, we presented a deep learning algorithm for AMC, inspired by residual learning, which has remarkable accuracy and great representational ability. We also employed a squeeze-and-excitation network that is capable of exploiting modeling interconnections between channels and adaptively re-calibrates the channel-wise feature response to improve performance. The proposed network was designed to meet the accuracy requirements with a reduced number of parameters for efficiency. The proposed model was evaluated on two benchmark datasets and compared with existing methods. The results show that the proposed model outperforms existing methods in terms of accuracy and has up to 72.5% fewer parameters than convolutional neural network designs.

Machine Learning-Based Error Correction Codes and Communication Protocols for Power Line Communication: An Overview

“…This process facilitates the autoencoder in discovering a robust representation of the input signal, thereby enabling the identification of optimal encoding and decoding strategies for stochastic channels. Notably, the autoencoder can achieve a solution that surpasses the performance of existing modulation and encoding methods [160,161]. Moreover, the autoencoder approach operates without making any assumptions about the channel and, in theory, can comprehend the dynamics of the PLC channel without relying on its current manifestation as a linear periodic timevariable system [67,160].…”

Terahertz Meets AI: The State of the Art

“…One way to apply RL for THz in ns-3 is to optimize the parameters of a THz communication system, such as transmit power, modulation [186], and beamforming, based on the current channel conditions. For example, the RL algorithm can learn from past interactions with the simulated environment and adjust the system parameters to maximize a particular performance metric, such as data rate or capacity.…”