Scalable End-to-End RF Classification: A Case Study on Undersized Dataset Regularization by Convolutional-MST
This addresses the problem of high data acquisition costs in RF sensing for applications like radar and communications, though it appears incremental as it builds on existing deep learning techniques.
The paper tackles the challenge of limited data in RF signal classification by introducing a new deep learning approach based on multistage training, achieving over 99% accuracy for up to 17 classes with only 11 samples per class, which is a 35% improvement over standard methods.
Unlike areas such as computer vision and speech recognition where convolutional and recurrent neural networks-based approaches have proven effective to the nature of the respective areas of application, deep learning (DL) still lacks a general approach suitable for the unique nature and challenges of RF systems such as radar, signals intelligence, electronic warfare, and communications. Existing approaches face problems in robustness, consistency, efficiency, repeatability and scalability. One of the main challenges in RF sensing such as radar target identification is the difficulty and cost of obtaining data. Hundreds to thousands of samples per class are typically used when training for classifying signals into 2 to 12 classes with reported accuracy ranging from 87% to 99%, where accuracy generally decreases with more classes added. In this paper, we present a new DL approach based on multistage training and demonstrate it on RF sensing signal classification. We consistently achieve over 99% accuracy for up to 17 diverse classes using only 11 samples per class for training, yielding up to 35% improvement in accuracy over standard DL approaches.