Optimal Feature Selection With Cryptographic Process for IDs based Cloud Environment

In today's cloud computing landscape, ensuring robust security measures is paramount to safeguarding sensitive data and infrastructure from cyber threats. This paper proposes a novel approach, the Hashing ECC-HMM (HECC-HMM), for enhancing cloud security through intrusion detection. By combining the principles of Hashing and Error-Correcting Codes (ECC) with the Hidden Markov Model (HMM), the HECC-HMM offers a powerful framework for accurately detecting and classifying security threats in cloud environments. To evaluate the effectiveness of the HECC-HMM, extensive experiments are conducted using two widely used datasets: CICIDS and KDD. The results demonstrate the HECC-HMM's ability to achieve a detection accuracy of 98.5% on the CICIDS dataset and 96.7% on the KDD dataset, with corresponding false positive rates of 1.2% and 1.5%, and false negative rates of 2.3% and 3.2% respectively. Furthermore, the HECC-HMM exhibits efficient real-time intrusion detection capabilities, with a computational overhead of only 0.5 seconds per sample for CICIDS and 0.8 seconds per sample for KDD. These numerical results underscore the efficacy of the HECC-HMM approach in enhancing cloud security, providing a reliable and efficient means of mitigating risks and safeguarding cloud infrastructure and data from evolving cyber threats.