Scholarlink

<p>In recent years, cloud computing has emerged as a centralized storage network among education and commercial users. To guarantee the access and security of heterogeneous devices, various approaches have been deployed that resolve the potential issues however faced adverse setbacks concerning key generation, as well as identifying malicious packets. Therefore, the research proposes a Lepidoptera Stepwise movement Optimized hybrid attention-enabled light gradient boosting machine with a bi-directional long short-term memory (LeSMO-ALB2TM) framework for addressing the downsides in attack detection. The Lepidoptera step-wise optimized Synthetic minority sampling (LeSM-SyOT) is introduced for preventing class imbalances and user authentication is facilitated through the signature generation using the Elgamal scheme. Moreover, the secure storage of authenticated data packets is expedited through the Lattice-based signature standard (LatS2 scheme) that improves data integrity with a minimum storage of 432.17 Kilo bytes. Moreover, the Lepidoptera Stepwise movement optimization (LeSMO) algorithm is employed for hyperparameter tuning which enhances detection and sampling. The experimental results reveal the superiority with a higher accuracy of 97.98% in attack detection and higher encryption-decryption rates of 0.64 and 0.75.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

LeSMO-ALB2TM: Data security in educational cloud using lattice signature scheme and optimized deep learning framework for attack detection

  • 期刊论文

  • 发表日期：

    2025年9月15日

查看全文
  • Kavuri KSVA Satheesh,
  • T. krishnaSree

摘要

In recent years, cloud computing has emerged as a centralized storage network among education and commercial users. To guarantee the access and security of heterogeneous devices, various approaches have been deployed that resolve the potential issues however faced adverse setbacks concerning key generation, as well as identifying malicious packets. Therefore, the research proposes a Lepidoptera Stepwise movement Optimized hybrid attention-enabled light gradient boosting machine with a bi-directional long short-term memory (LeSMO-ALB2TM) framework for addressing the downsides in attack detection. The Lepidoptera step-wise optimized Synthetic minority sampling (LeSM-SyOT) is introduced for preventing class imbalances and user authentication is facilitated through the signature generation using the Elgamal scheme. Moreover, the secure storage of authenticated data packets is expedited through the Lattice-based signature standard (LatS2 scheme) that improves data integrity with a minimum storage of 432.17 Kilo bytes. Moreover, the Lepidoptera Stepwise movement optimization (LeSMO) algorithm is employed for hyperparameter tuning which enhances detection and sampling. The experimental results reveal the superiority with a higher accuracy of 97.98% in attack detection and higher encryption-decryption rates of 0.64 and 0.75.