5924626f19493a390957f07 20250125063458172 beie:beie director@blueeyesintelligence.org WEB-FORM International Journal of Innovative Technology and Exploring Engineering IJITEE 22783075 10.35940/ijitee https://www.ijitee.org/ 01 30 2025 14 2 School of Computing, SASTRA Deemed-to-be University, Thanjavur (Tamil Nadu), India. Dr. Venkata Subramanian N. https://orcid.org/0000-0001-8096-742X Dr. Shankar Sriram V S. https://orcid.org/0000-0001-7870-7944 Center for Information Super Highway (CISH), School of Computing, SASTRA Deemed-to-be University, Thanjavur (Tamil Nadu), India. Safeguarding virtual machines (VMs) during migration is essential to avert Service Level Agreement (SLA) violations. This research article presents a robust security framework that utilizes deep learning and advanced encryption methods to reduce the impact of Distributed Denial of Service (DDoS) attacks during virtual machine migration. The study introduces an Improved Sparrow Search Algorithm-based Deep Neural Network (ISSA-DNN) for the classification of DDoS attacks and utilizes Advanced Encryption Standard-Elliptic Curve Cryptography (AES-ECC) to safeguard virtual machine images. The primary objective is to mitigate the risks associated with VM migration by identifying DDoS attacks and safeguarding VMs using advanced cryptographic techniques. The research employs the Canadian Institute for Cybersecurity Distributed Denial of Service (CICDDoS) dataset, implementing preprocessing procedures like duplication elimination, feature selection via Random Forest, and normalization to improve the precision of the DNN classifier. The ISSA-DNN approach enhances hyperparameter optimization by inverse mutation-based sparrow search, yielding a precise attack classification model. Furthermore, the research incorporates AES-ECC for encrypting VM images, amalgamating AES's computational efficiency with ECCs improved security. In contrast to conventional methods, this hybrid encryption approach enhances throughput and decreases encryption and decryption durations, rendering it appropriate for high-throughput and real-time applications. Experimental findings indicate that the proposed ISSA-DNN attains a classification accuracy of 98.79%, surpassing current state-of-the-art techniques. The AES-ECC encryption technique markedly enhances performance metrics, safeguarding the security of virtual machines during migration. This proactive security policy safeguards sensitive data and guarantees adherence to regulatory standards. In conclusion, the established framework offers a comprehensive solution for mitigating DDoS attacks and safeguarding VM migration via advanced deep learning and encryption methodologies. Integrating ISSA-DNN for attack classification and AES-ECC for encryption offers a robust strategy for improving cybersecurity in cloud environments. 01 30 2025 01 30 2025 12 20 CC-BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijitee.org true International Journal of Innovative Technology and Exploring Engineering (IJITEE) Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP) Based on my understanding, this article has no conflicts of interest. This article has not been sponsored or funded by any organization or agency. The independence of this research is a crucial factor in affirming its impartiality, as it has been conducted without any external sway. The data provided in this article is exempt from the requirement for ethical approval or participant consent. The adequate resources of this article are publicly accessible. The authorship of this article is contributed equally to all participating individuals. 10.35940/ijitee.B1032.14020125 https://www.ijitee.org/portfolio-item/B103214020125/