A Two-Stage Time-Domain Equalization Method for Mitigating Nonlinear Distortion in Single-Carrier THz Communication Systems

Terahertz (THz) communication is regarded as a key technology for achieving high-speed data transmission and wireless communication due to its ultra-high frequency and large bandwidth characteristics. In this study, we focus on a single-carrier THz communication system and propose a two-stage deep l...

Full description

Saved in:
Bibliographic Details
Main Authors: Yunchuan Liu, Hongcheng Yang, Ziqi Liu, Minghan Jia, Shang Li, Jiajie Li, Jingsuo He, Zhe Yang, Cunlin Zhang
Format: Article
Language:English
Published: MDPI AG 2025-08-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/15/4825
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849406162972180480
author Yunchuan Liu
Hongcheng Yang
Ziqi Liu
Minghan Jia
Shang Li
Jiajie Li
Jingsuo He
Zhe Yang
Cunlin Zhang
author_facet Yunchuan Liu
Hongcheng Yang
Ziqi Liu
Minghan Jia
Shang Li
Jiajie Li
Jingsuo He
Zhe Yang
Cunlin Zhang
author_sort Yunchuan Liu
collection DOAJ
description Terahertz (THz) communication is regarded as a key technology for achieving high-speed data transmission and wireless communication due to its ultra-high frequency and large bandwidth characteristics. In this study, we focus on a single-carrier THz communication system and propose a two-stage deep learning-based time-domain equalization method, specifically designed to mitigate the nonlinear distortions in such systems, thereby enhancing communication reliability and performance. The method adopts a progressive learning strategy, whereby global characteristics are initially captured before progressing to local levels. This enables the effective identification and equalization of channel characteristics, particularly in the mitigation of nonlinear distortion and random interference, which can otherwise negatively impact communication quality. In an experimental setting at a frequency of 230 GHz and a channel distance of 2.1 m, this method demonstrated a substantial reduction in the system’s bit error rate (BER), exhibiting particularly noteworthy performance enhancements in comparison to before equalization. To validate the model’s generalization capability, data collection and testing were also conducted at a frequency of 310 GHz and a channel distance of 1.5 m. Experimental results show that the proposed time-domain equalizer, trained using the two-stage DL framework, achieved significant BER reductions of approximately 92.15% at 230 GHz (2.1 m) and 83.33% at 310 GHz (1.5 m), compared to the system’s performance prior to equalization. The method exhibits stable performance under varying conditions, supporting its use in future THz communication studies.
format Article
id doaj-art-0ba34784d1a8443aa32b3dbaa41f05a0
institution Kabale University
issn 1424-8220
language English
publishDate 2025-08-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-0ba34784d1a8443aa32b3dbaa41f05a02025-08-20T03:36:30ZengMDPI AGSensors1424-82202025-08-012515482510.3390/s25154825A Two-Stage Time-Domain Equalization Method for Mitigating Nonlinear Distortion in Single-Carrier THz Communication SystemsYunchuan Liu0Hongcheng Yang1Ziqi Liu2Minghan Jia3Shang Li4Jiajie Li5Jingsuo He6Zhe Yang7Cunlin Zhang8Key Laboratory of Terahertz Optoelectronics, Ministry of Education, Beijing 100048, ChinaKey Laboratory of Terahertz Optoelectronics, Ministry of Education, Beijing 100048, ChinaKey Laboratory of Terahertz Optoelectronics, Ministry of Education, Beijing 100048, ChinaKey Laboratory of Terahertz Optoelectronics, Ministry of Education, Beijing 100048, ChinaKey Laboratory of Terahertz Optoelectronics, Ministry of Education, Beijing 100048, ChinaKey Laboratory of Terahertz Optoelectronics, Ministry of Education, Beijing 100048, ChinaKey Laboratory of Terahertz Optoelectronics, Ministry of Education, Beijing 100048, ChinaKey Laboratory of Terahertz Optoelectronics, Ministry of Education, Beijing 100048, ChinaKey Laboratory of Terahertz Optoelectronics, Ministry of Education, Beijing 100048, ChinaTerahertz (THz) communication is regarded as a key technology for achieving high-speed data transmission and wireless communication due to its ultra-high frequency and large bandwidth characteristics. In this study, we focus on a single-carrier THz communication system and propose a two-stage deep learning-based time-domain equalization method, specifically designed to mitigate the nonlinear distortions in such systems, thereby enhancing communication reliability and performance. The method adopts a progressive learning strategy, whereby global characteristics are initially captured before progressing to local levels. This enables the effective identification and equalization of channel characteristics, particularly in the mitigation of nonlinear distortion and random interference, which can otherwise negatively impact communication quality. In an experimental setting at a frequency of 230 GHz and a channel distance of 2.1 m, this method demonstrated a substantial reduction in the system’s bit error rate (BER), exhibiting particularly noteworthy performance enhancements in comparison to before equalization. To validate the model’s generalization capability, data collection and testing were also conducted at a frequency of 310 GHz and a channel distance of 1.5 m. Experimental results show that the proposed time-domain equalizer, trained using the two-stage DL framework, achieved significant BER reductions of approximately 92.15% at 230 GHz (2.1 m) and 83.33% at 310 GHz (1.5 m), compared to the system’s performance prior to equalization. The method exhibits stable performance under varying conditions, supporting its use in future THz communication studies.https://www.mdpi.com/1424-8220/25/15/4825deep learningterahertzdigital communicationLSTM
spellingShingle Yunchuan Liu
Hongcheng Yang
Ziqi Liu
Minghan Jia
Shang Li
Jiajie Li
Jingsuo He
Zhe Yang
Cunlin Zhang
A Two-Stage Time-Domain Equalization Method for Mitigating Nonlinear Distortion in Single-Carrier THz Communication Systems
Sensors
deep learning
terahertz
digital communication
LSTM
title A Two-Stage Time-Domain Equalization Method for Mitigating Nonlinear Distortion in Single-Carrier THz Communication Systems
title_full A Two-Stage Time-Domain Equalization Method for Mitigating Nonlinear Distortion in Single-Carrier THz Communication Systems
title_fullStr A Two-Stage Time-Domain Equalization Method for Mitigating Nonlinear Distortion in Single-Carrier THz Communication Systems
title_full_unstemmed A Two-Stage Time-Domain Equalization Method for Mitigating Nonlinear Distortion in Single-Carrier THz Communication Systems
title_short A Two-Stage Time-Domain Equalization Method for Mitigating Nonlinear Distortion in Single-Carrier THz Communication Systems
title_sort two stage time domain equalization method for mitigating nonlinear distortion in single carrier thz communication systems
topic deep learning
terahertz
digital communication
LSTM
url https://www.mdpi.com/1424-8220/25/15/4825
work_keys_str_mv AT yunchuanliu atwostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT hongchengyang atwostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT ziqiliu atwostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT minghanjia atwostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT shangli atwostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT jiajieli atwostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT jingsuohe atwostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT zheyang atwostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT cunlinzhang atwostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT yunchuanliu twostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT hongchengyang twostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT ziqiliu twostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT minghanjia twostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT shangli twostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT jiajieli twostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT jingsuohe twostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT zheyang twostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems
AT cunlinzhang twostagetimedomainequalizationmethodformitigatingnonlineardistortioninsinglecarrierthzcommunicationsystems