Spectrum sensing based on adversarial transfer learning

Abstract Recently, deep learning (DL) based spectrum sensing (SS) has drawn much attention due to its better capacity of feature extraction and superb performance. However, the model robustness of the DL based scheme is limited by reason of the dynamic radio environment, leading to the floating of s...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiawu Miao, Yuebo Li, Xiaojun Jing, Fangpei Zhang, Junsheng Mu
Format: Article
Language:English
Published: Wiley 2022-10-01
Series:IET Communications
Online Access:https://doi.org/10.1049/cmu2.12459
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849392035549675520
author Jiawu Miao
Yuebo Li
Xiaojun Jing
Fangpei Zhang
Junsheng Mu
author_facet Jiawu Miao
Yuebo Li
Xiaojun Jing
Fangpei Zhang
Junsheng Mu
author_sort Jiawu Miao
collection DOAJ
description Abstract Recently, deep learning (DL) based spectrum sensing (SS) has drawn much attention due to its better capacity of feature extraction and superb performance. However, the model robustness of the DL based scheme is limited by reason of the dynamic radio environment, leading to the floating of sensing performance. Motivated by this, adversarial transfer learning is applied to SS here, where the model is pre‐trained at the central node firstly and fine‐tuned at the local nodes. More specifically, a 2D dataset of the observed signal is constructed under various signal‐to‐noise‐ratio (SNRs) and a convolution neural network (CNN) model is designed. Then a part of samples with various SNRs in the constructed dataset are employed to pre‐train the proposed CNN model. After that, the pre‐trained CNN model is distributed to local nodes with different SNRs and the pre‐trained CNN model is fine‐tuned. The proposed CNN model is pre‐trained based on the samples under various SNRs, resulting in its stronger adaptability at the local node. The simulation experiments validate the effectiveness of the proposed scheme.
format Article
id doaj-art-19450948e8cc4295a6cfaec8c7b8ddaf
institution Kabale University
issn 1751-8628
1751-8636
language English
publishDate 2022-10-01
publisher Wiley
record_format Article
series IET Communications
spelling doaj-art-19450948e8cc4295a6cfaec8c7b8ddaf2025-08-20T03:40:51ZengWileyIET Communications1751-86281751-86362022-10-0116172059206910.1049/cmu2.12459Spectrum sensing based on adversarial transfer learningJiawu Miao0Yuebo Li1Xiaojun Jing2Fangpei Zhang3Junsheng Mu4School of information and Communication Engineering Beijing University of Posts and Telecommunications Beijing People's Republic of ChinaSchool of information and Communication Engineering Beijing University of Posts and Telecommunications Beijing People's Republic of ChinaSchool of information and Communication Engineering Beijing University of Posts and Telecommunications Beijing People's Republic of ChinaInformation Science Academy of China Electronics Technology Group Corporation Beijing People's Republic of ChinaSchool of information and Communication Engineering Beijing University of Posts and Telecommunications Beijing People's Republic of ChinaAbstract Recently, deep learning (DL) based spectrum sensing (SS) has drawn much attention due to its better capacity of feature extraction and superb performance. However, the model robustness of the DL based scheme is limited by reason of the dynamic radio environment, leading to the floating of sensing performance. Motivated by this, adversarial transfer learning is applied to SS here, where the model is pre‐trained at the central node firstly and fine‐tuned at the local nodes. More specifically, a 2D dataset of the observed signal is constructed under various signal‐to‐noise‐ratio (SNRs) and a convolution neural network (CNN) model is designed. Then a part of samples with various SNRs in the constructed dataset are employed to pre‐train the proposed CNN model. After that, the pre‐trained CNN model is distributed to local nodes with different SNRs and the pre‐trained CNN model is fine‐tuned. The proposed CNN model is pre‐trained based on the samples under various SNRs, resulting in its stronger adaptability at the local node. The simulation experiments validate the effectiveness of the proposed scheme.https://doi.org/10.1049/cmu2.12459
spellingShingle Jiawu Miao
Yuebo Li
Xiaojun Jing
Fangpei Zhang
Junsheng Mu
Spectrum sensing based on adversarial transfer learning
IET Communications
title Spectrum sensing based on adversarial transfer learning
title_full Spectrum sensing based on adversarial transfer learning
title_fullStr Spectrum sensing based on adversarial transfer learning
title_full_unstemmed Spectrum sensing based on adversarial transfer learning
title_short Spectrum sensing based on adversarial transfer learning
title_sort spectrum sensing based on adversarial transfer learning
url https://doi.org/10.1049/cmu2.12459
work_keys_str_mv AT jiawumiao spectrumsensingbasedonadversarialtransferlearning
AT yueboli spectrumsensingbasedonadversarialtransferlearning
AT xiaojunjing spectrumsensingbasedonadversarialtransferlearning
AT fangpeizhang spectrumsensingbasedonadversarialtransferlearning
AT junshengmu spectrumsensingbasedonadversarialtransferlearning