Optimization algorithms in security and privacy-preserving data disturbance for collaborative edge computing social IoT deep learning architectures

Abstract

Abstract Edge computing (EC) has been viewed as a viable option for addressing cloud computing's shortcomings in supporting Internet-of-Things (IoT) applications as an innovative strategy. However, edge-assisted IoT security requires to be better guaranteed due to the network's instability and increase in attack surfaces. Practice of safeguarding a given network from threats that could compromise its availability is referred to as network security. In addition, we can add that network security must address issues such as unauthorized access to network-accessible resources and even their misuse. This study proposes a novel approach to improving edge computing-based data privacy through secure data transmission and deep learning-based optimization with Internet of Things. Edge network privacy preservation is accomplished through collaborative architecture and hybrid federated sever-based stochastic vector networks. The trust based multiple encryption algorithm is used to secure the transmission of the data after it has been optimized using firefly grey optimization. Throughput, network security analysis, PDR, latency, and energy consumption were all examined as part of the experimental analysis for network data privacy and optimization.