{"title":"QAE-IDS: DDoS anomaly detection in IoT devices using Post-Quantization Training","authors":"B. S. Sharmila, Rohini Nagapadma","doi":"10.1080/23080477.2023.2260023","DOIUrl":null,"url":null,"abstract":"ABSTRACTOver the past few years, many intellectuals have focused on unsupervised learning for anomaly detection in IoT networks. Deploying an unsupervised Autoencoder algorithm for Intrusion Detection System (IDS) is computationally intensive for IoT devices with limited resources. In this work, we propose two distinct AI models using Post-Training Quantization; Quantized Autoencoder float16 (QAE-float16) and Quantized Autoencoder uint8 (QAE-uint8). QAE models are derived using Autoencoder models, which work on the assumption of generating high Reconstruction Error (RE) for anomaly data. Post Training Quantization includes pruning, clustering, and Quantization techniques. The proposed models were tested against the RT-IoT23 dataset, which includes normal and attack traces. This study is focused on the three major types of attacks, namely SSH brute-force, UFONet and DDoS (Distributed Denial of Service) exploitation. Since these attacks are the gateway for future exploitation. The model performance evaluated on IoT devices reveals that QAE-uint8 is the most efficient model by a wide margin, with average memory utilization decreased by 70.01%, memory size compressed by 92.23%, and peak CPU utilization decreased by 27.94%. Therefore, the proposed QAE-uint8 model has the potential to be used in low-power IoT Edge devices to detect anomalies.KEYWORDS: Anomaly detectionartificial intelligenceautoencodersIoTIDSpost-quantization training Disclosure statementNo potential conflit of interest was reported by the authors.","PeriodicalId":53436,"journal":{"name":"Smart Science","volume":"12 1","pages":"0"},"PeriodicalIF":2.4000,"publicationDate":"2023-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Smart Science","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1080/23080477.2023.2260023","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"MULTIDISCIPLINARY SCIENCES","Score":null,"Total":0}
引用次数: 0
Abstract
ABSTRACTOver the past few years, many intellectuals have focused on unsupervised learning for anomaly detection in IoT networks. Deploying an unsupervised Autoencoder algorithm for Intrusion Detection System (IDS) is computationally intensive for IoT devices with limited resources. In this work, we propose two distinct AI models using Post-Training Quantization; Quantized Autoencoder float16 (QAE-float16) and Quantized Autoencoder uint8 (QAE-uint8). QAE models are derived using Autoencoder models, which work on the assumption of generating high Reconstruction Error (RE) for anomaly data. Post Training Quantization includes pruning, clustering, and Quantization techniques. The proposed models were tested against the RT-IoT23 dataset, which includes normal and attack traces. This study is focused on the three major types of attacks, namely SSH brute-force, UFONet and DDoS (Distributed Denial of Service) exploitation. Since these attacks are the gateway for future exploitation. The model performance evaluated on IoT devices reveals that QAE-uint8 is the most efficient model by a wide margin, with average memory utilization decreased by 70.01%, memory size compressed by 92.23%, and peak CPU utilization decreased by 27.94%. Therefore, the proposed QAE-uint8 model has the potential to be used in low-power IoT Edge devices to detect anomalies.KEYWORDS: Anomaly detectionartificial intelligenceautoencodersIoTIDSpost-quantization training Disclosure statementNo potential conflit of interest was reported by the authors.
期刊介绍:
Smart Science (ISSN 2308-0477) is an international, peer-reviewed journal that publishes significant original scientific researches, and reviews and analyses of current research and science policy. We welcome submissions of high quality papers from all fields of science and from any source. Articles of an interdisciplinary nature are particularly welcomed. Smart Science aims to be among the top multidisciplinary journals covering a broad spectrum of smart topics in the fields of materials science, chemistry, physics, engineering, medicine, and biology. Smart Science is currently focusing on the topics of Smart Manufacturing (CPS, IoT and AI) for Industry 4.0, Smart Energy and Smart Chemistry and Materials. Other specific research areas covered by the journal include, but are not limited to: 1. Smart Science in the Future 2. Smart Manufacturing: -Cyber-Physical System (CPS) -Internet of Things (IoT) and Internet of Brain (IoB) -Artificial Intelligence -Smart Computing -Smart Design/Machine -Smart Sensing -Smart Information and Networks 3. Smart Energy and Thermal/Fluidic Science 4. Smart Chemistry and Materials