{"title":"基于无监督dl的无线信号异常检测","authors":"Xiangli Liu, Wei Tan, Zan Li, Junjie Zeng","doi":"10.1016/j.dsp.2025.105578","DOIUrl":null,"url":null,"abstract":"<div><div>Detecting wireless signal anomalies in non-ideal channels and complex electromagnetic environments is a particularly challenging and demanding task. Susceptible to environmental influences, wireless signal anomalies are diverse, making its anomaly detection very difficult. To improve the detection of anomalous signals in complex electromagnetic environments, a novel unsupervised model, CAAEDS (Combined Adversarial Autoencoder and Deep SVDD) is proposed, which incorporates dual input modalities: time-domain data and Power Spectral Density (PSD). CAAEDS extracts time domain data feature information using Long Short-term Memory (LSTM) and PSD data feature information using Residual Networks (ResNets). Results from experiments demonstrate that: 1) The performance of the proposed algorithm outperforms state-of-the-art algorithms. 2) Ablation studies prove that CAAEDS can overcome the shortcomings of unsupervised AAE and Deep SVDD in wireless signal anomaly detection. 3) Wireless signal datasets collected in real-world environments verify the ability of CAAEDS to adapt to the environment and detect weak anomalies in wireless signals.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105578"},"PeriodicalIF":3.0000,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Unsupervised DL-based wireless signal anomaly detection\",\"authors\":\"Xiangli Liu, Wei Tan, Zan Li, Junjie Zeng\",\"doi\":\"10.1016/j.dsp.2025.105578\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Detecting wireless signal anomalies in non-ideal channels and complex electromagnetic environments is a particularly challenging and demanding task. Susceptible to environmental influences, wireless signal anomalies are diverse, making its anomaly detection very difficult. To improve the detection of anomalous signals in complex electromagnetic environments, a novel unsupervised model, CAAEDS (Combined Adversarial Autoencoder and Deep SVDD) is proposed, which incorporates dual input modalities: time-domain data and Power Spectral Density (PSD). CAAEDS extracts time domain data feature information using Long Short-term Memory (LSTM) and PSD data feature information using Residual Networks (ResNets). Results from experiments demonstrate that: 1) The performance of the proposed algorithm outperforms state-of-the-art algorithms. 2) Ablation studies prove that CAAEDS can overcome the shortcomings of unsupervised AAE and Deep SVDD in wireless signal anomaly detection. 3) Wireless signal datasets collected in real-world environments verify the ability of CAAEDS to adapt to the environment and detect weak anomalies in wireless signals.</div></div>\",\"PeriodicalId\":51011,\"journal\":{\"name\":\"Digital Signal Processing\",\"volume\":\"168 \",\"pages\":\"Article 105578\"},\"PeriodicalIF\":3.0000,\"publicationDate\":\"2025-09-09\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Digital Signal Processing\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S1051200425006001\",\"RegionNum\":3,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ENGINEERING, ELECTRICAL & ELECTRONIC\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Digital Signal Processing","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1051200425006001","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
引用次数: 0
摘要
在非理想信道和复杂电磁环境中检测无线信号异常是一项特别具有挑战性和高要求的任务。无线信号异常多种多样,易受环境影响,异常检测难度很大。为了提高复杂电磁环境下异常信号的检测能力,提出了一种新的无监督模型CAAEDS (Combined Adversarial Autoencoder and Deep SVDD),该模型结合了时域数据和功率谱密度(PSD)的双输入模式。CAAEDS利用LSTM (Long - Short-term Memory)提取时域数据特征信息,利用ResNets (Residual Networks)提取PSD数据特征信息。实验结果表明:1)该算法的性能优于现有算法。2)消融研究证明,CAAEDS可以克服无监督AAE和Deep SVDD在无线信号异常检测中的不足。3)在实际环境中采集的无线信号数据集验证了CAAEDS对环境的适应能力和检测无线信号中微弱异常的能力。
Unsupervised DL-based wireless signal anomaly detection
Detecting wireless signal anomalies in non-ideal channels and complex electromagnetic environments is a particularly challenging and demanding task. Susceptible to environmental influences, wireless signal anomalies are diverse, making its anomaly detection very difficult. To improve the detection of anomalous signals in complex electromagnetic environments, a novel unsupervised model, CAAEDS (Combined Adversarial Autoencoder and Deep SVDD) is proposed, which incorporates dual input modalities: time-domain data and Power Spectral Density (PSD). CAAEDS extracts time domain data feature information using Long Short-term Memory (LSTM) and PSD data feature information using Residual Networks (ResNets). Results from experiments demonstrate that: 1) The performance of the proposed algorithm outperforms state-of-the-art algorithms. 2) Ablation studies prove that CAAEDS can overcome the shortcomings of unsupervised AAE and Deep SVDD in wireless signal anomaly detection. 3) Wireless signal datasets collected in real-world environments verify the ability of CAAEDS to adapt to the environment and detect weak anomalies in wireless signals.
期刊介绍:
Digital Signal Processing: A Review Journal is one of the oldest and most established journals in the field of signal processing yet it aims to be the most innovative. The Journal invites top quality research articles at the frontiers of research in all aspects of signal processing. Our objective is to provide a platform for the publication of ground-breaking research in signal processing with both academic and industrial appeal.
The journal has a special emphasis on statistical signal processing methodology such as Bayesian signal processing, and encourages articles on emerging applications of signal processing such as:
• big data• machine learning• internet of things• information security• systems biology and computational biology,• financial time series analysis,• autonomous vehicles,• quantum computing,• neuromorphic engineering,• human-computer interaction and intelligent user interfaces,• environmental signal processing,• geophysical signal processing including seismic signal processing,• chemioinformatics and bioinformatics,• audio, visual and performance arts,• disaster management and prevention,• renewable energy,