Intrusion Detection in IoT Using Deep Learning
Abstract
:1. Introduction
2. Related Work
3. Material and Methods
3.1. Step 1: Bot-IoT
3.2. Step 2: Pre-Processing of the Datasets
3.3. Step 3: Feature Selection
3.4. Step 4: Classification
3.5. Step 5: Trained, Tested, and Evaluated
4. Results
5. Conclusions
6. Future Works
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Zhang, J.; Pan, L.; Han, Q.-L.; Chen, C.; Wen, S.; Xiang, Y. Deep learning-based attack detection for cyber-physical system cybersecurity: A survey. IEEE/CAA J. Autom. Sin. 2021, 9, 377–391. [Google Scholar] [CrossRef]
- Lee, I. Internet of Things (IoT) cybersecurity: Literature review and IoT cyber risk management. Futur. Internet 2020, 12, 157. [Google Scholar] [CrossRef]
- Almiani, M.; AbuGhazleh, A.; Al-Rahayfeh, A.; Atiewi, S.; Razaque, A. Deep recurrent neural network for IoT intrusion detection system. Simul. Model. Pract. Theory 2020, 101, 102031. [Google Scholar] [CrossRef]
- Azumah, S.W.; Elsayed, N.; Adewopo, V.; Zaghloul, Z.S.; Li, C. A deep lstm based approach for intrusion detection iot devices network in smart home. In Proceedings of the 2021 IEEE 7th World Forum on Internet of Things (WF-IoT), New Orleans, LA, USA, 26–31 July 2021. [Google Scholar]
- Thakkar, A.; Lohiya, R. A review on machine learning and deep learning perspectives of IDS for IoT: Recent updates, security issues, and challenges. Arch. Comput. Methods Eng. 2021, 28, 3211–3243. [Google Scholar] [CrossRef]
- Li, Y.; Zuo, Y.; Song, H.; Lv, Z. Deep learning in security of internet of things. IEEE Internet Things J. 2021; early access. [Google Scholar] [CrossRef]
- Idrissi, I.; Boukabous, M.; Azizi, M.; Moussaoui, O.; El Fadili, H. Toward a deep learning-based intrusion detection system for IoT against botnet attacks. IAES Int. J. Artif. Intell. (IJ-AI) 2021, 10, 110. [Google Scholar] [CrossRef]
- Venkatraman, S.; Surendiran, B. Adaptive hybrid intrusion detection system for crowd sourced multimedia internet of things systems. Multimedia Tools Appl. 2019, 79, 3993–4010. [Google Scholar] [CrossRef]
- Alladi, T.; Chamola, V.; Sikdar, B.; Choo, K.-K.R. Consumer IoT: Security vulnerability case studies and solutions. IEEE Consum. Electron. Mag. 2020, 9, 17–25. [Google Scholar] [CrossRef]
- Asharf, J.; Moustafa, N.; Khurshid, H.; Debie, E.; Haider, W.; Wahab, A. A Review of Intrusion Detection Systems Using Machine and Deep Learning in Internet of Things: Challenges, Solutions and Future Directions. Electronics 2020, 9, 1177. [Google Scholar] [CrossRef]
- Wang, X.; Zhao, Y.; Pourpanah, F. Recent advances in deep learning. Int. J. Mach. Learn. Cybern. 2020, 11, 747–750. [Google Scholar] [CrossRef]
- Abu Al-Haija, Q.; Zein-Sabatto, S. An efficient deep-learning-based detection and classification system for cyber-attacks in IoT communication networks. Electronics 2020, 9, 2152. [Google Scholar] [CrossRef]
- Abu Al-Haija, Q.; Al-Dala’ien, M.A. ELBA-IoT: An Ensemble Learning Model for Botnet Attack Detection in IoT Networks. J. Sens. Actuator Netw. 2022, 11, 18. [Google Scholar] [CrossRef]
- Pioneering Deep Learning in the Cyber Security Space: The New Standard? Information Age. 25 March 2020. Available online: https://www.information-age.com/pioneering-deep-learning-cyber-security-new-standard-123488524/ (accessed on 9 November 2020).
- Khan, T.; Sarkar, R.; Mollah, A.F. Deep learning approaches to scene text detection: A comprehensive review. Artif. Intell. Rev. 2021, 54, 3239–3298. [Google Scholar] [CrossRef]
- Aversano, L.; Bernardi, M.L.; Cimitile, M.; Pecori, R. A systematic review on Deep Learning approaches for IoT security. Comput. Sci. Rev. 2021, 40, 100389. [Google Scholar] [CrossRef]
- Stefanos, T.; Lagkas, T.; Rantos, K. Deep learning in iot intrusion detection. J. Netw. Syst. Manag. 2022, 30, 1–40. [Google Scholar]
- Davis, B.D.; Mason, J.C.; Anwar, M. Mason, and Mohd Anwar. Vulnerability studies and security postures of IoT devices: A smart home case study. IEEE Internet Things J. 2020, 7, 10102–10110. [Google Scholar] [CrossRef]
- Jiang, X.; Lora, M.; Chattopadhyay, S. An experimental analysis of security vulnerabilities in industrial IoT devices. ACM Trans. Internet Technol. 2020, 20, 1–24. [Google Scholar] [CrossRef]
- Chanal, P.M.; Kakkasageri, M.S. Kakkasageri. Security and privacy in IOT: A survey. Wirel. Pers. Commun. 2020, 115, 1667–1693. [Google Scholar] [CrossRef]
- Susilo, B.; Sari, R.F. Intrusion Detection in IoT Networks Using Deep Learning Algorithm. Information 2020, 11, 279. [Google Scholar] [CrossRef]
- Ibitoye, O.; Shafiq, O.; Matrawy, A. Analyzing Adversarial Attacks against Deep Learning for Intrusion Detection in IoT Networks. In Proceedings of the 2019 IEEE Global Communications Conference (GLOBECOM), Waikoloa, HI, USA, 9–13 December 2019; pp. 1–6. [Google Scholar] [CrossRef] [Green Version]
- Ge, M.; Fu, X.; Syed, N.; Baig, Z.; Teo, G.; Robles-Kelly, A. Deep Learning-Based Intrusion Detection for IoT Networks. In Proceedings of the 2019 IEEE 24th Pacific Rim International Symposium on Dependable Computing (PRDC), Kyoto, Japan, 1–3 December 2019; pp. 256–25609. [Google Scholar] [CrossRef]
- Alkadi, O.; Moustafa, N.; Turnbull, B.; Choo, K.-K.R. A Deep Blockchain Framework-enabled Collaborative Intrusion Detection for Protecting IoT and Cloud Networks. IEEE Internet Things J. 2020, 8, 1. [Google Scholar] [CrossRef]
- Parra, G.D.L.T.; Rad, P.; Choo, K.-K.R.; Beebe, N. Detecting Internet of Things attacks using distributed deep learning. J. Netw. Comput. Appl. 2020, 163, 102662. [Google Scholar] [CrossRef]
- Samy, A.; Yu, H.; Zhang, H. Fog-Based Attack Detection Framework for Internet of Things Using Deep Learning. IEEE Access 2020, 8, 74571–74585. [Google Scholar] [CrossRef]
- Pecori, R.; Tayebi, A.; Vannucci, A.; Veltri, L. IoT Attack Detection with Deep Learning Analysis. In Proceedings of the 2020 International Joint Conference on Neural Networks (IJCNN), Glasgow, UK, 19–24 July 2020; pp. 1–8. [Google Scholar] [CrossRef]
- Farsi, M. Application of ensemble RNN deep neural network to the fall detection through IoT environment. Alex. Eng. J. 2021, 60, 199–211. [Google Scholar] [CrossRef]
- Shobana, M.; Poonkuzhali, S. A novel approach to detect IoT malware by system calls using Deep learning techniques. In Proceedings of the 2020 International Conference on Innovative Trends in Information Technology (ICITIIT), Kottayam, India, 13–14 February 2020; pp. 1–5. [Google Scholar] [CrossRef]
- Koroniotis, N.; Moustafa, N.; Sitnikova, E.; Turnbull, B. Towards the development of realistic botnet dataset in the internet of things for network forensic analytics: Bot-iot dataset. Future Gener. Comput. Syst. 2019, 100, 779–796. [Google Scholar] [CrossRef] [Green Version]
- Koroniotis, N.; Moustafa, N.; Sitnikova, E.; Slay, J. Towards developing network forensic mechanism for botnet activities in the IoT based on machine learning techniques. In Proceedings of the International Conference on Mobile Networks and Management, Melbourne, Australia, 13–15 December 2017; Springer: Cham, Switzerland, 2017. [Google Scholar]
- Koroniotis, N. Designing an effective network forensic framework for the investigation of botnets in the Internet of Things. Ph.D. Dissertation, UNSW Sydney, Sydney, Australia, 2020. [Google Scholar]
- Koroniotis, N.; Moustafa, N.; Schiliro, F.; Gauravaram, P.; Janicke, H. A holistic review of cybersecurity and reliability perspectives in smart airports. IEEE Access 2020, 8, 209802–209834. [Google Scholar] [CrossRef]
- Koroniotis, N.; Moustafa, N. Enhancing network forensics with particle swarm and deep learning: The particle deep framework. arXiv 2020, arXiv:2005.00722. [Google Scholar]
- Peterson, J.M.; Leevy, J.L.; Khoshgoftaar, T.M. A review and analysis of the bot-iot dataset. In Proceedings of the 2021 IEEE International Conference on Service-Oriented System Engineering (SOSE), Oxford, UK, 23–26 August 2021. [Google Scholar]
Ref. | Classifier | Dataset | Accuracy and Performance | Limitation |
---|---|---|---|---|
[21] | Convolutional neural network (CNN) | Bot-IoT | The accuracy achieved 91.27% in batch size 128 and the lowest 88.30% in batch size 32. | The accuracy decreases when using 32 and 64 batch size. |
[22] | Feedforward Neural Networks (FNN) | Bot-IoT | The accuracy achieved 95.1%, and the average precision, recall, and F1-score reached 95%. | The Bot-IoT dataset’s feature normalization shows that the accuracy would drop below 50%. |
[23] | Feed-forward neural networks (FNN) | Bot-IoT | The accuracy achieved 99.414% in multi-class classification for DDoS/DoS attacks and 99% across all evaluation measures: accuracy, precision, recall, and F1 score. | The proposed solution has proved to be less precise in protecting against keylogging attacks and information theft in binary classification, also, the multi-class classification has achieved the low accuracy of 88.9%. |
[24] | Bidirectional Long Short-Term Memory (BiLSTM) | Bot-IoT and UNSW-NB15 | The accuracy achieved 98.91 %, the detection rate achieved 99.79% in the Bot-IoT dataset, the accuracy has reached 99.41%, and the detection rate achieved 99.95% in the UNSW-NB15 dataset. | The proposed solution’s limitation is that IDS’ performance degrades under heavy network traffic and underperforms in alarming against and detecting a complex attack. |
[25] | Long short-term memory model (LSTM) | N_BaIoT | The accuracy achieved 97.84% at the back-end level and precision achieved 97.81%, recall 95%, F-score 96.25%, FPR 0.0001, and TPR 0.9999 | The proposed solution does not function on detecting emerging attacks. |
[26] | Long short-term memory model (LSTM) | N_BaIoT-2018 , CICIDS-2017, RPLNIDS-2017 and NSL-KDD | The accuracy achieved in binary classification 99.85% in N_BaIoT 2018 dataset and precision achieved 98.64%, recalled 99.81%, F-score 99.12%, FPR 0.1%, and DR 99.81%. | The proposed model needs massive datasets and longer time to train. |
[27] | Deep Neural Network (DNN) | 4 datasets of IoT | The accuracy achieved in binary classification 99.75% with seven hidden layers and precision achieved 99.37%, recalled 99.37%, and F-score 99.37% | The experiment tests on only four different attacks (Scanning, DoS, MITM, and Mirai). |
[28] | Long short-term memory model (LSTM) Additionally, Random Forest (RF) | SmartFall dataset | The accuracy achieved in LSTM is 93.4% and precision achieved 92%, recalled 93.4%, and F-score 91.78%. The accuracy achieved in RF 99.9%. Additionally, precision achieved 99.9%, recalled 99.9%, and F-score 99.9%. | The accuracy of LSTM is considered low compared with other methods and techniques. The study applied only one dataset as an evaluation. |
[29] | Recurrent neural network (RNN) | IOTPOT | The accuracy achieved is 98.712%. Additionally, error rate is 1.288%. | It could be improved by implementing a multiclass classification and category malware on system calls. It could also be enhanced by applying other deep learning methods, such as LSTM. |
Number | Parameter | Explanation |
---|---|---|
1 | Classifier | CNN |
2 | Layers | (4) Input (10) Hidden and (1) Output |
3 | Input Features | 4 |
4 | Output | Normal (0) Attack (1) |
5 | Training Dataset | 80% for Training 20% for Testing |
Number | Parameter | Explanation |
---|---|---|
1 | Classifier | LSTM Additionally, GRU |
2 | Layers | (4) Input (100) Hidden and (1) Output |
3 | Input Features | 4 |
4 | Output | Normal (0) Attack (1) |
5 | Training Dataset | 80% for Training 20% for Testing |
Classifier | Dataset | Accuracy | FA |
---|---|---|---|
CNN | Bot-IoT | 0.997 | 0.003 |
LSTM | Bot-IoT | 0.998 | 0.002 |
GRU | Bot-IoT | 0.996 | 0.004 |
Classifier | Dataset | Precision | Recall | F1 Score |
---|---|---|---|---|
CNN | Bot-IoT | 0.996 | 0.999 | 0.998 |
LSTM | Bot-IoT | 0.997 | 1.000 | 0.998 |
GRU | Bot-IoT | 0.996 | 1.000 | 0.998 |
Ref | Classifier | Dataset | Accuracy and Performance |
---|---|---|---|
[21] | Convolutional Neural Network (CNN) | Bot-IoT | The accuracy achieves 91.27% in batch size 128 and the lowest 88.30% in batch size 32. |
[22] | Feedforward Neural Networks (FNN) | Bot-IoT | The accuracy achieves 95.1%, and the average precision, recall, and F1-score reached 0.95% |
[24] | Bidirectional Long Short-Term Memory (BiLSTM) | Bot-IoT | The accuracy achieves 98.91 %, and the detection rate achieved 99.79% |
Proposed model | Convolutional Neural Network (CNN) | Bot-IoT | The accuracy achieves 99.7% and precision achieve 99.6%, recall 99.9%, and F-score and detection rate 99.8% |
Proposed model | Long Short-Term Memory (LSTM) | Bot-IoT | The accuracy achieves 99.8%, precision achieve 99.7%, recall 100%, and F-score and detection rate 99.8% |
Proposed model | Gated Recurrent Unit (GRU) | Bot-IoT | The accuracy achieves 99.6% and precision achieve 99.6%, recall 100%, and F-score and detection rate 99.8% |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Banaamah, A.M.; Ahmad, I. Intrusion Detection in IoT Using Deep Learning. Sensors 2022, 22, 8417. https://doi.org/10.3390/s22218417
Banaamah AM, Ahmad I. Intrusion Detection in IoT Using Deep Learning. Sensors. 2022; 22(21):8417. https://doi.org/10.3390/s22218417
Chicago/Turabian StyleBanaamah, Alaa Mohammed, and Iftikhar Ahmad. 2022. "Intrusion Detection in IoT Using Deep Learning" Sensors 22, no. 21: 8417. https://doi.org/10.3390/s22218417
APA StyleBanaamah, A. M., & Ahmad, I. (2022). Intrusion Detection in IoT Using Deep Learning. Sensors, 22(21), 8417. https://doi.org/10.3390/s22218417