Supervised Contrastive Learning for Uncertainty-Aware Wireless Signal Analysis: A Case Study for Modulation Classification
- 7 April 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 74 (00906778) , 7227-7240
Abstract
Artificial intelligence (AI), and more specifically deep learning techniques, have demonstrated strong capabilities in processing wireless signals, enabling automatic modulation classification (AMC). However, existing AI-based AMC methods often produce unreliable predictions and lack robustness to out-of-distribution (OOD) inputs, which limits their deployment in real-world scenarios. This study aims to address the gap in simulations to real deployments by fitting predictions to OOD scenarios. We propose an uncertainty-aware AMC framework based on supervised contrastive learning (SupCon). The proposed framework aims to enhance classification reliability, particularly under OOD conditions. In this framework, a ResNet-based representation learning model that consists of a feature extractor and a projection head is first trained using a combination of SupCon loss and cross-entropy (CE) loss to produce class-discriminative embeddings. By decoupling the shared representation learning model from task-specific classifiers, the framework enables a modular and computationally efficient uncertainty estimation strategy, avoiding redundant computation during inference. The learned embeddings are then processed by a set of class-wise binary classifiers that provide both classification and sample-wise uncertainty estimates. A rejection mechanism is incorporated to improve decision reliability by discarding uncertain predictions. We evaluate the proposed framework on the RadioML 2018 dataset. Experimental results show that our approach significantly improves representation quality and classification reliability compared to conventional supervised learning. Specifically, the classification accuracy increases from 63.7% to 93.6% under in-distribution conditions, and from 30.8% to 92.5% under 50% OOD contamination, while maintaining over 85% recall on accepted predictions.Keywords
Funding Information
- MITACS and Ericsson Canada, and NSERC Canada Research Chairs program
This publication has 26 references indexed in Scilit:
- Multi-Modal Data-Enhanced Foundation Models for Prediction and Control in Wireless Networks: A SurveyIEEE Communications Surveys & Tutorials, 2025
- Generalized Out-of-Distribution Detection: A SurveyInternational Journal of Computer Vision, 2024
- Relaxed Contrastive Learning for Federated Learning2024
- Domain Generalization in Machine Learning Models for Wireless Communications: Concepts, State-of-the-Art, and Open IssuesIEEE Communications Surveys & Tutorials, 2023
- Artificial Intelligence-Driven Real-Time Automatic Modulation Classification Scheme for Next-Generation Cellular NetworksIEEE Access, 2021
- Automatic Modulation Classification: A Deep Architecture SurveyIEEE Access, 2021
- Intelligent Radio Signal Processing: A SurveyIEEE Access, 2021
- A survey of traditional and advanced automatic modulation classification techniques, challenges, and some novel trendsInternational Journal of Communication Systems, 2021
- Big Data Processing Architecture for Radio Signals Empowered by Deep Learning: Concept, Experiment, Applications and ChallengesIEEE Access, 2018
- FaceNet: A unified embedding for face recognition and clustering2015