Semantic RF Waveform Adaptation in Flying Ad Hoc Networks via Hybrid Knowledge Bases and Deep Reinforcement Learning

By directly encoding the semantic meaning into each RF waveform symbol, we can significantly reduce the communication overhead. Particularly, semantic communication in Flying Ad Hoc Networks (FANETs)—formed by uncrewed aerial vehicles (UAVs)—is a promising application for enabl...

Full description

Saved in:
Bibliographic Details
Main Authors: Weiqiang Lyu, Linsheng He, Jiamiao Zhao, Fei Hu
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11021564/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849332580775624704
author Weiqiang Lyu
Linsheng He
Jiamiao Zhao
Fei Hu
author_facet Weiqiang Lyu
Linsheng He
Jiamiao Zhao
Fei Hu
author_sort Weiqiang Lyu
collection DOAJ
description By directly encoding the semantic meaning into each RF waveform symbol, we can significantly reduce the communication overhead. Particularly, semantic communication in Flying Ad Hoc Networks (FANETs)—formed by uncrewed aerial vehicles (UAVs)—is a promising application for enabling efficient airborne networks. This paper introduces a semantics-aware approach to optimize OFDM (Orthogonal Frequency Division Multiplexing) waveforms, leveraging Deep Q-Network (DQN)-based reinforcement learning (RL) with nonlinear optimization, enhanced by an innovative hybrid knowledge base shared by all nodes. Our method incorporates semantic importance through adaptive parameter tuning and constrained optimization. The hybrid knowledge base allows for efficient representation and utilization of semantic information. We propose a framework that jointly optimizes modulation order and cyclic prefix length while considering the semantic importance of transmitted data. The system is formulated as a Markov Decision Process (MDP) with scalable state/action spaces. This algorithm enhances the system’s ability to handle diverse and dynamic communication scenarios, adapting to varying semantic importance levels.
format Article
id doaj-art-175d8f8aa2ad4f8ab4ea97bd00e69571
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-175d8f8aa2ad4f8ab4ea97bd00e695712025-08-20T03:46:09ZengIEEEIEEE Access2169-35362025-01-0113981909820710.1109/ACCESS.2025.357614611021564Semantic RF Waveform Adaptation in Flying Ad Hoc Networks via Hybrid Knowledge Bases and Deep Reinforcement LearningWeiqiang Lyu0https://orcid.org/0009-0002-7187-4592Linsheng He1https://orcid.org/0000-0001-9067-5121Jiamiao Zhao2https://orcid.org/0000-0002-3670-219XFei Hu3https://orcid.org/0000-0003-4346-9477Department of Electrical and Computer Engineering, The University of Alabama, Tuscaloosa, AL, USADepartment of Electrical and Computer Engineering, The University of Alabama, Tuscaloosa, AL, USADepartment of Electrical and Computer Engineering, The University of Alabama, Tuscaloosa, AL, USADepartment of Electrical and Computer Engineering, The University of Alabama, Tuscaloosa, AL, USABy directly encoding the semantic meaning into each RF waveform symbol, we can significantly reduce the communication overhead. Particularly, semantic communication in Flying Ad Hoc Networks (FANETs)—formed by uncrewed aerial vehicles (UAVs)—is a promising application for enabling efficient airborne networks. This paper introduces a semantics-aware approach to optimize OFDM (Orthogonal Frequency Division Multiplexing) waveforms, leveraging Deep Q-Network (DQN)-based reinforcement learning (RL) with nonlinear optimization, enhanced by an innovative hybrid knowledge base shared by all nodes. Our method incorporates semantic importance through adaptive parameter tuning and constrained optimization. The hybrid knowledge base allows for efficient representation and utilization of semantic information. We propose a framework that jointly optimizes modulation order and cyclic prefix length while considering the semantic importance of transmitted data. The system is formulated as a Markov Decision Process (MDP) with scalable state/action spaces. This algorithm enhances the system’s ability to handle diverse and dynamic communication scenarios, adapting to varying semantic importance levels.https://ieeexplore.ieee.org/document/11021564/Cyclic prefixdeep reinforcement learninghybrid knowledge basemodulationOFDMsemantic communications
spellingShingle Weiqiang Lyu
Linsheng He
Jiamiao Zhao
Fei Hu
Semantic RF Waveform Adaptation in Flying Ad Hoc Networks via Hybrid Knowledge Bases and Deep Reinforcement Learning
IEEE Access
Cyclic prefix
deep reinforcement learning
hybrid knowledge base
modulation
OFDM
semantic communications
title Semantic RF Waveform Adaptation in Flying Ad Hoc Networks via Hybrid Knowledge Bases and Deep Reinforcement Learning
title_full Semantic RF Waveform Adaptation in Flying Ad Hoc Networks via Hybrid Knowledge Bases and Deep Reinforcement Learning
title_fullStr Semantic RF Waveform Adaptation in Flying Ad Hoc Networks via Hybrid Knowledge Bases and Deep Reinforcement Learning
title_full_unstemmed Semantic RF Waveform Adaptation in Flying Ad Hoc Networks via Hybrid Knowledge Bases and Deep Reinforcement Learning
title_short Semantic RF Waveform Adaptation in Flying Ad Hoc Networks via Hybrid Knowledge Bases and Deep Reinforcement Learning
title_sort semantic rf waveform adaptation in flying ad hoc networks via hybrid knowledge bases and deep reinforcement learning
topic Cyclic prefix
deep reinforcement learning
hybrid knowledge base
modulation
OFDM
semantic communications
url https://ieeexplore.ieee.org/document/11021564/
work_keys_str_mv AT weiqianglyu semanticrfwaveformadaptationinflyingadhocnetworksviahybridknowledgebasesanddeepreinforcementlearning
AT linshenghe semanticrfwaveformadaptationinflyingadhocnetworksviahybridknowledgebasesanddeepreinforcementlearning
AT jiamiaozhao semanticrfwaveformadaptationinflyingadhocnetworksviahybridknowledgebasesanddeepreinforcementlearning
AT feihu semanticrfwaveformadaptationinflyingadhocnetworksviahybridknowledgebasesanddeepreinforcementlearning