بهبود دقت سیستم تشخیص نفوذ در اینترنت اشیا با استفاده از الگوریتم‌های یادگیری ماشین و خوشه‌بندی
محورهای موضوعی : پردازش چند رسانه ای، سیستمهای ارتباطی، سیستمهای هوشمند
جواد پاشایی باربین
1
,
مهدی جلالی
2
1 - استادیار، گروه کامپیوتر، واحد نقده، دانشگاه آزاد اسلامی، نقده، ایران
2 - استادیار، گروه برق، دانشکده فنی مهندسی، واحد نقده، دانشگاه آزاد اسلامی، نقده، ایران
کلید واژه: تشخیص نفوذ, یادگیری ماشین, داده کاوی, ماشین بردار پشتیبان, k-means,
چکیده مقاله :
افزایش استفاده از اینترنت اشیا منجر به افزایش حملات در این شبکه ها شده است. سازندگان دستگاه های اینترنت اشیا علاقه مند به کاهش هزینه ها با نادیده گرفتن مقررات امنیتی هستند که باعث آسیب گسترده و مانع از رشد اینترنت اشیا میشود. گسترش حملات مبتنی بر اینترنت اشیا تا زمانی ادامه خواهدداشت که سازندگان اینترنت اشیا مکانیزم‌های پاسخگویی و امنیتی را در دستگاه‌های خود بگنجانند. تا آن زمان، اینترنت اشیا این پتانسیل را دارد که به محیطی برای حملات سایبری آینده تبدیل شود که چالشهای بزرگی را به همراه خواهدداشت. از این رو، در این تحقیق راهکارهای برقراری امنیت اینترنت اشیا بررسی شده و راه حلی مبتنی بر ترکیب ماشین بردار پشتیبان و الگوریتم k-means ارائه شده است. نتایج نشان میدهد که دقت روش پیشنهادی 98.35 درصد است که کارآمدی روش پیشنهادی را نشان میدهد و قابلیت پیاده سازی برای تشخیص خطا به صورت عملی را دارد.
Abstract
Introduction: The recent rise of the Internet of Things (IoT) has led to increasing attacks in IoT. Manufacturers of IoT devices are interested in reducing costs by ignoring security regulations that cause widespread damage and impede the growth of the IoT. The proliferation of IoT-based attacks will continue as long as IoT manufacturers incorporate accountability and security mechanisms into their devices. The proliferation of IoT-based attacks will continue as long as IoT manufacturers incorporate accountability and security mechanisms into their devices. Until then, the Internet of Things has the potential to become an environment for future cyber-attacks, which will pose great challenges.
Method: In this research, the solutions for establishing security in the Internet of Things have been investigated and have provided a solution based on the combination of support vector machine and K-means algorithm. First, preprocessing is applied to the data set and the data that has no effect on the result are deleted. Then, the support vector machine algorithm is applied to the data set and the intrusion or non-intrusion status is determined. This proposed method achieves better results by applying k-means to the data set, and the combination of support vector machine algorithms and k-means improves the accuracy of the proposed method.
Results: The results showed that the proposed method is more efficient than previous methods. this study sought to improve the security challenge in wireless sensor networks. The proposed method of this research is to use a combination of support vector machine and chi-mean, which showed very good performance compared to previous methods. According to the studies and the proposed method, it can be found that the best method in detecting and detecting intrusion is the use of K-Means algorithm, which can be achieved with 98.35% accuracy using the support vector machine method and K-Means algorithm.
Discussion: The most important criterion for determining the performance of an algorithm is the Accuracy criterion. This criterion calculates the total accuracy of a category. This criterion indicates what percentage of the total data set is properly categorized. This criterion is the evaluation based on the accuracy and the accuracy of the proposed method is better than the previously presented methods.
[1] M. Bauer, and J.W. Walewski, “The IoT Architectural Reference Model as Enabler, in Enabling Things to Talk,” 2013, Springer. p. 17-25.
[2] E. De Coninck, , et al. , “Distributed neural networks for Internet of Things,” the Big-Little approach. in International Internet of Things Summit. 2015. Springer.
[3] J.Cañedo, and A. Skjellum. , “Using machine learning to secure IoT systems. in Privacy, Security and Trust (PST),” 2016 14th Annual Conference on. 2016. IEEE.
[4] S. Mansfield-Devine, “The growth and evolution of DDoS” Network Security, 2015. 2015(10): p. 13-20.
[5] M. Nazarpour, N. Nezafati, , S. Shokouhyar, 'Using the Modified Colonial Competition Algorithm to Increase the Speed and Accuracy of the Intelligent Intrusion Detection System', Intelligent Multimedia Processing and Communication Systems (IMPCS),2023, 4(1), pp. 1-10. [Persian].
[6] V. Adat, and B. Gupta, “Security in Internet of Things: issues, challenges, taxonomy, and architecture” Telecommunication Systems, 2018. 67(3): p. 423-441.
[7] S. Hanif, T. Ilyas and M. Zeeshan, “Intrusion Detection In IoT Using Artificial Neural Networks On UNSW-15 Dataset,” 2019 IEEE 16th International Conference on Smart Cities: Improving Quality of Life Using ICT & IoT and AI (HONET-ICT), 2019, pp. 152-156, doi: 10.1109/HONET.2019.8908122.
[8] M. A. Rahman, et al, “Scalable machine learning-based intrusion detection system for IoT-enabled smart cities,” Sustainable Cities and Society 61: 102324.
[9] N. Chaabouni, M. Mosbah, A. Zemmari and C. Sauvignac, “A OneM2M Intrusion Detection and Prevention System based on Edge Machine Learning,” NOMS 2020 - 2020 IEEE/IFIP Network Operations and Management Symposium, 2020, pp. 1-7, doi: 10.1109/NOMS47738.2020.9110473.
[10] A. Aldaej, “Enhancing Cyber Security in Modern Internet of things (IoT) Using Intrusion Prevention Algorithm for IoT (IPAI),” in IEEE Access, doi: 10.1109/ACCESS.2019.2893445.
[11] M. Akhlaghpour, 'Providing a Solution Based on Fuzzy Logic to Reduce False Positive Alarms in The Intrusion Detection System', Intelligent Multimedia Processing and Communication Systems (IMPCS), 2(4),2021, pp. 45-50. [Persian].
[12] R.S. Nayaki, and A.S. Kumar, “An Analysis of DDoS Attack Detection and Mitigation Using Machine Learning System,” International Journal on Recent and Innovation Trends in Computing and Communication, 2017. 5(10): p. 80-82.
[13] C. Yin, et al., “A deep learning approach for intrusion detection using recurrent neural networks;” IEEE Access, 2017. 5: p. 21954-21961.
[14] M. Safaldin, M. Otair, and L. Abualigah, “Improved binary gray wolf optimizer and SVM for intrusion detection system in wireless sensor networks,” J Ambient Intell Human Comput 12, 1559–1576 (2021). https://doi.org/10.1007/s12652-020-02228-z
