[Retracted] Secure Big Data Processing in Multihoming Networks with AI‐Enabled IoT

Abstract

The constant investigation, as well as dispensation of data among various processing, has been influenced by computerized strategies enabled by artificial neural network associated with Internet of Things, as well as cloud-dependent organizations. Multihoming is also a category of an organization that brings together several categories of organizations in its atmosphere during the dealing with various information. These days, the massive amounts of information being handled also observed in multihoming systems are given less thought, reducing the security risk and effectiveness of data processing and observation. The application of ANN-associated frameworks among multihoming massive information using IoT as well as AI-associated frameworks could be beneficial in a variety of ways. The constant investigation and dispensation of information among various processing are influenced by computerized strategies enabled by ANN associated with IoT, as well as cloud-associated organizations. Multihoming is also a kind of organization that brings together several kinds of organizations into a unique situation while dealing with a large volume of information. When multihoming enormous amounts of data with IoT and AI-integrated systems, there are various advantages to employing AI-based solutions. Despite the fact that multihoming security difficulties and their evaluation have been carefully explored by various scientists and researchers, big data security analysis in multihoming, particularly when using automated methodologies and systems, receives little attention. These days, the massive amounts of information being handled and observed in multihoming systems are given less thought, reducing the security risk and effectiveness of data processing and observation. The application of AI-based frameworks in multihoming massive information with IoT associated with AI frameworks could be beneficial in a variety of ways. Even though multihoming security challenges and their analysis have been widely focused on by many researchers and specialists, little attention has been paid to them.

Privacy-Preserving Big Data Security for IoT With Federated Learning and Cryptography

“…Our proposed approach surpassed both existing approaches in communication overhead performance. Notably, our proposed approach achieved a reduction of up to 35% in the amount of data transmitted compared to [27] and up to 44% compared to [31]. These findings serve as evidence of the efficacy of our proposed approach in minimizing communication overhead while upholding the same level of accuracy.…”

“…The authors use the integration of data processing and cryptographic techniques to show how this strategy can improve the security and effectiveness of IoT big data systems. Venu et al [31] propose a framework that combines AI techniques with network layer security controls to ensure effective and secure data processing in IoT environments. The availability, confidentiality, and integrity of data in multi-homing networks are just a few of the security issues that this method addresses in IoT big data systems.…”

“…We conducted experiments to evaluate the effectiveness of our suggested approach using the widely adopted FedSim simulator for federated learning system assessment. Our approach was compared against two other methods proposed by Stergiou et al [27] and Venu et al [31]. In the experiment, we utilized a simulation setup comprising of 1000 IoT devices, each generating data samples of size 1000.…”

“…In this section, we assess the accuracy of our proposed federated learning method using the FedSim simulator and compare it with two existing techniques, [27] and [31]. We replicate a scenario where 100 IoT devices are dispersed across 10 clusters, each cluster comprising of 10 devices.…”

“…In this section, we compare the resource usage of the proposed approach to that of [27] and [31]. We monitor 50 IoT devices' CPU and memory usage throughout the federated learning process, which lasts for 10 iterations with a batch size of 10.…”