A malicious feature detection and prevention mechanism with BRDH approach for improved security in Homomorphic Blockchain

【Author】 Swanthana, K.; Aravinth, S. S.

【Source】KNOWLEDGE-BASED SYSTEMS

【影响因子】8.139

【Abstract】In today's digital landscape, safeguarding sensitive data from external threats is a pressing concern. This paper introduces an innovative approach named Buffalo Recurrent Diffie Hellman (BRDH), designed to bolster data security by proactively identifying and mitigating potential risks. BRDH combines the power of African Buffalo optimization and Recurrent Neural Networks (RNN) to continuously monitor data in the cloud, swiftly detecting and responding to suspicious activities. Data encryption is employed to establish robust security measures using the Diffie-Hellman algorithm. Homomorphic encryption further enhances security by validating hash values and denying access in case of discrepancies. Authorized users receive a private key to decrypt cypher Text and access the original data securely. BRDH significantly enhances data confidentiality, achieves over 99 % accuracy in malicious feature detection, and reduces encryption/decryption times compared to conventional models, as demonstrated through applications in the stock market, healthcare, and network traffic data. BRDH sets a new standard in robust data protection amidst evolving security threats. Advantages

【Keywords】Buffalo recurrent Diffie Hellman; African Buffalo optimization; Recurrent Neural Networks (RNN); Diffie Hellman algorithm; Data encryption; Homomorphic; Decrypt cypher Text

【发表时间】2025 FEB 15

【收录时间】2025-04-07

【文献类型】

【主题类别】

--