Efficient Training of Deep Classifiers for Wireless Source Identification Using Test SNR Estimates

We study efficient deep learning training algorithms that process received wireless signals, if a test Signal to Noise Ratio (SNR) estimate is available. We focus on two tasks that facilitate source identification: 1- Identifying the modulation type, 2- Identifying the wireless technology and channe...

Full description

Saved in:
Bibliographic Details
Published inIEEE wireless communications letters Vol. 9; no. 8; pp. 1314 - 1318
Main Authors Wang, Xingchen, Ju, Shengtai, Zhang, Xiwen, Ramjee, Sharan, Gamal, Aly El
Format Journal Article
LanguageEnglish
Published Piscataway IEEE 01.08.2020
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects
Accuracy
Algorithms
Boosting
channel identification
Computer architecture
Deep learning
Deep learning for wireless
Greedy algorithms
Kernel
Machine learning
Modulation
modulation classification
Signal processing
Signal to noise ratio
SNR bagging
SNR boosting
Training
Wireless communication
Online AccessGet full text

Cover

More Information
Summary:We study efficient deep learning training algorithms that process received wireless signals, if a test Signal to Noise Ratio (SNR) estimate is available. We focus on two tasks that facilitate source identification: 1- Identifying the modulation type, 2- Identifying the wireless technology and channel in the 2.4 GHz ISM band. For benchmarking, we rely on recent literature on testing deep learning algorithms against two well-known datasets. We first demonstrate that using training data corresponding only to the test SNR value leads to dramatic reductions in training time while incurring a small loss in average test accuracy, as it improves the accuracy for low SNR values. Further, we show that an erroneous test SNR estimate with a small positive offset is better for training than another having the same error magnitude with a negative offset. Secondly, we introduce a greedy training SNR Boosting algorithm that leads to uniform improvement in accuracy across all tested SNR values, while using a small subset of training SNR values at each test SNR. Finally, we demonstrate the potential of bootstrap aggregating (Bagging) based on training SNR values to improve generalization at low test SNR values with scarcity of training data.
ISSN:2162-2337
2162-2345
DOI:10.1109/LWC.2020.2989286