A comprehensive deep learning benchmark for IoT IDS

被引:34
作者
Ahmad, Rasheed [1 ]
Alsmadi, Izzat [2 ]
Alhamdani, Wasim [1 ]
Tawalbeh, Lo'ai [2 ]
机构
[1] Univ Cumberlands, 6178 Coll Stn Dr, Williamsburg, KY 40769 USA
[2] Univ Texas A&M San Antonio, One Univ Way, San Antonio, TX 78224 USA
关键词
Intrusion detection system (IDS); Machine learning; Deep learning; Large-scale attacks; Internet of Things (IoT); Benchmark network dataset; NETWORK INTRUSION DETECTION; ATTACK DETECTION; DETECTION SYSTEM; INTERNET; THINGS; SURVEILLANCE; ANALYTICS; FRAMEWORK;
D O I
10.1016/j.cose.2021.102588
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The significance of an intrusion detection system (IDS) in networks security cannot be overstated in detecting and responding to malicious attacks. Failure to detect large-scale attacks like DDoS not only makes the networks vulnerable, but a failure of critical lifesaving medical and industrial equipment can also put human lives at risk. Lack of availability of comprehensive and quality network datasets and the narrow scope to build an IDS based on a single machine learning classifier adds further limitations. Such issues can risk producing inaccurate or biased results in the solutions proposed by various researchers. Toward this end, this paper analyzed several datasets (old, recent, non-IoT, and IoT specific) using several individual and hybrid deep learning classifiers. Our goal is to establish a benchmark that can compare several classification models on several datasets to limit (1) dataset quality issues and (2) possible bias in produced results. We reported our empirical results by revealing exciting findings on some of the classifiers, which took hours to converge but could not successfully detect attacks. In contrast, others quickly converged and were able to produce the best results in terms of accuracy and other performance metrics. We believe that this paper's findings will help build a comprehensive IDS by recognizing that classification or prediction models should be trained beyond a limited scope of one dataset or application. (C) 2021 Elsevier Ltd. All rights reserved.
引用
收藏
页数:22
相关论文
共 100 条
  • [71] Deep Learning for IoT Big Data and Streaming Analytics: A Survey
    Mohammadi, Mehdi
    Al-Fuqaha, Ala
    Sorour, Sameh
    Guizani, Mohsen
    [J]. IEEE COMMUNICATIONS SURVEYS AND TUTORIALS, 2018, 20 (04): : 2923 - 2960
  • [72] Cyber Attacks Detection based on Deep Learning for Cloud-Dew Computing in Automotive IoT Applications
    Moussa, Mohamed Mounir
    Alazzawi, Lubna
    [J]. 2020 IEEE INTERNATIONAL CONFERENCE ON SMART CLOUD (SMARTCLOUD 2020), 2020, : 55 - 61
  • [73] Moustafa N, 2015, 2015 MILITARY COMMUNICATIONS AND INFORMATION SYSTEMS CONFERENCE (MILCIS)
  • [74] An Ensemble Intrusion Detection Technique Based on Proposed Statistical Flow Features for Protecting Network Traffic of Internet of Things
    Moustafa, Nour
    Turnbull, Benjamin
    Choo, Kim-Kwang Raymond
    [J]. IEEE INTERNET OF THINGS JOURNAL, 2019, 6 (03): : 4815 - 4830
  • [75] Nagisetty A, 2019, PROCEEDINGS OF THE 2019 3RD INTERNATIONAL CONFERENCE ON COMPUTING METHODOLOGIES AND COMMUNICATION (ICCMC 2019), P633, DOI [10.1109/ICCMC.2019.8819688, 10.1109/iccmc.2019.8819688]
  • [76] Narla SRK, 2019, ITE J, V89, P28
  • [77] Naveed K, 2020, 2020 IFIP NETWORKING CONFERENCE AND WORKSHOPS (NETWORKING), P649
  • [78] Ng W., 2019, Soil Discuss, P1, DOI [10.5194/soil-2019-48, DOI 10.5194/SOIL-2019-48]
  • [79] DL-IDS: a deep learning-based intrusion detection framework for securing IoT
    Otoum, Yazan
    Liu, Dandan
    Nayak, Amiya
    [J]. TRANSACTIONS ON EMERGING TELECOMMUNICATIONS TECHNOLOGIES, 2022, 33 (03)
  • [80] Deep Learning for Encrypted Traffic Classification: An Overview
    Rezaei, Shahbaz
    Liu, Xin
    [J]. IEEE COMMUNICATIONS MAGAZINE, 2019, 57 (05) : 76 - 81