Edge Learning Based Collaborative Automatic Modulation Classification for Hierarchical Cognitive Radio Networks

TL;DR

This work tackles automatic modulation classification in hierarchical cognitive radio networks by leveraging edge learning to split the task between an edge device and an edge server. The proposed C-AMC framework combines a spectrum semantics compression network (SSCNet) at the edge with a modulation classification network (MCNet) at the server, enabling reduced transmission load and enhanced data privacy while maintaining high classification accuracy. The authors detail offline and online training procedures, and introduce model compression via weight pruning and quantization to fit the constraints of edge devices, accompanied by a complexity analysis. Simulation on the RadioML dataset demonstrates that C-AMC outperforms baselines in accuracy and efficiency, and provides practical insights for real-world deployment, including robustness to compression and scenario mismatch. The framework paves the way for scalable, privacy-preserving spectrum cognition in large-scale hierarchical CR systems.

Abstract

In hierarchical cognitive radio networks, edge or cloud servers utilize the data collected by edge devices for modulation classification, which, however, is faced with problems of the computation load, transmission overhead, and data privacy. In this article, an edge learning (EL) based framework jointly mobilizing the edge device and the edge server for intelligent co-inference is proposed to realize the collaborative automatic modulation classification (C-AMC) between them. A spectrum semantic compression neural network is designed for the edge device to compress the collected raw data into a compact semantic embedding that is then sent to the edge server via the wireless channel. On the edge server side, a modulation classification neural network combining the bidirectional long-short term memory and attention structures is elaborated to determine the modulation type from the noisy semantic embedding. The C-AMC framework decently balances the computation resources of both sides while avoiding the high transmission overhead and data privacy leakage. Both the offline and online training procedures of the C-AMC framework are elaborated. The compression strategy of the C-AMC framework is also developed to further facilitate the deployment, especially for the resource-constrained edge device. Simulation results show the superiority of the EL-based C-AMC framework in terms of the classification accuracy, computational complexity, and the data compression rate as well as reveal useful insights paving the practical implementation.

Figures (14)

• Figure 1: Hierarchical CR network model.
• Figure 2: EL-based C-AMC framework.
• Figure 3: Architecture of SSCNet and MCNet.
• Figure 4: Offline training procedure of C-AMC framework.
• Figure 5: Online training procedure of C-AMC framework.