عنوان مقاله:

نویسندگان:

رضا معمریامی، علی اکبر خزایی، سعید راحتی قوچانی

This research proposes a hybrid approach combining supervised machine learning algorithms with the VIKOR multi-criteria decision-making technique to enhance self-healing and fault management in next-generation networks. The primary challenges addressed include data imbalance, fault prioritization, and the complexity of managing heterogeneous networks. Three supervised learning algorithms—Naive Bayes, Decision Tree, and Random Forest—were integrated with VIKOR to tackle these issues. Results demonstrated that Random Forest achieved the highest accuracy (93.658%), but its Kappa score of zero indicated limitations in detecting minority classes. Decision Tree, with 92.6878% accuracy and the fastest runtime (0.04 seconds), proved to be the most suitable for real-time applications. Naive Bayes also performed well, with 91.143% accuracy and low prediction errors. The integration of VIKOR provided significant advantages, including fault prioritization based on severity and impact, improved detection of minority classes in imbalanced datasets, and multi-criteria decision-making for optimal resource management. This combination not only enhanced the accuracy and efficiency of the algorithms but also improved the system"s flexibility and scalability. Ultimately, the approach contributed to reduced response times, increased network reliability, and lower operational costs. The study highlights the effectiveness of combining machine learning techniques with multi-criteria decision-making methods like VIKOR to address the complex challenges of next-generation networks. By leveraging VIKOR"s ability to balance conflicting criteria and prioritize critical issues, the proposed framework offers a robust solution for fault detection, diagnosis, and compensation in dynamic and heterogeneous network environments. Future work could explore the application of this framework in other domains, such as IoT and edge computing, to further validate its adaptability and effectiveness.

Multi-criteria decision making, self-organizing networks, self-healing, management of defective cells, VIKOR

برای لینک‌دهی به این مقاله از آدرس زیر استفاده نمایید:

برای ارجاع در منابع از عبارت زیر استفاده کنید:
معمریامی، رضا، اکبر خزایی، علی، راحتی قوچانی، سعید، 1404، یادگیری ماشین و تکنیک ویکور برای تشخیص ناهنجاری و رتبه بندی سلول در 5G&B5G، بیست و یکمین کنفرانس ملی دانشجویی مهندسی برق ایران، https://conference.uok.ac.ir/fa/archive_index.php?c=HN-iscee21&aid=1517971