An Advancing Temporal Convolutional Network for 5G Latency Services via Automatic Modulation Recognition
- 18 February 2022
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 69 (6) , 3002-3006
Abstract
Automatic modulation recognition (AMR) has received significant attention since its decisive factor for modern non-cooperative communication systems. Meanwhile, the existing works on deep learning technique achieve exceptional accuracy; however, these works dissatisfy real-time requirements for 5G low-latency services. To remedy this flaw, this brief proposes a low-latency AMR method by applying temporal convolutional network (TCN). Furthermore, the principal component analysis (PCA)-based TCN and uniform subsampling-based TCN methods are leveraged to further alleviate the computation complexity and render real-time TCN viable. Experimental results demonstrate that the proposed method can achieve lower complexity and superior recognition accuracy than existing works and pave the way for 5G low-latency services.Keywords
Funding Information
- National Natural Science Foundation (61772099, 61772098)
This publication has 15 references indexed in Scilit:
- Fast Service Migration in 5G Trends and ScenariosIEEE Network, 2020
- Short Block-Length Codes for Ultra-Reliable Low Latency CommunicationsIEEE Communications Magazine, 2018
- Automatic Modulation Classification: A Deep Learning Enabled ApproachIEEE Transactions on Vehicular Technology, 2018
- Artificial Neural Network Based Automatic Modulation Classification over a Software Defined Radio Testbed2018
- Over-the-Air Deep Learning Based Radio Signal ClassificationIEEE Journal of Selected Topics in Signal Processing, 2018
- Online Hybrid Likelihood Based Modulation Classification Using Multiple SensorsIEEE Transactions on Wireless Communications, 2017
- Deep architectures for modulation recognition2017
- Convolutional Radio Modulation Recognition Networks2016
- Deep Residual Learning for Image Recognition2016
- A Bayesian Approach for Adaptively Modulated Signals Recognition in Next-Generation CommunicationsIEEE Transactions on Signal Processing, 2015