Federated AI based learning platform through protecting data privacy and security for heterogeneous data
Main Article Content
Abstract
This paper proposes a Federated Learning Secure Approach (FLSA) to enhance data privacy and security in IoT-based healthcare systems. By integrating encryption and privacy-preserving mechanisms, the framework ensures secure data processing and sharing, reducing latency and improving scalability. Performance analysis demonstrates the framework’s reliability in maintaining data accuracy while minimizing service overhead. Those data’s need reliable and secure communication as it is the details with more confidential data also as it helps to develop an automated and secure communication in the field of healthcare. In the advanced era, confidential data’s being generated from the IoT devices are deployed in distributed manner. In order to secure the user information’s, it needs strength user privacy policy as it helps to have customized network infrastructure and it is applicable for various advance application in real time especially for healthcare. To preserve the user policy, the federated learning will act as the machine learning to optimize the data being offered to deploy smart cities. In this paper, FL based Secure Approach is proposed which is enabled with federated learning and encryption policy, which is enabled with user privacy preservation especially for automated healthcare application. In this approach, security and privacy is ensured by applying the environment of encryption based on complex data in cloud. Instead of sending the user personal data to the cloud platform, this approach makes those user data to be learned using federated machine learning technology. In the performance analysis, service overhead is being calculated based on the probability of services offered and its service reliability is analysed. Thus, this proposed approach makes the latency to be reduced based on the storage requirement.