Abstract
Internet of things (IoT) has increased an importance for an area of interest in many devices. Then, the applications such as sensitive home sensors, medical devices, wireless sensors,and other devices are related to IoT network. The transmission of big data is subject to a possible attack that could cause network interruptions and problems with security. The security performance prediction is important for IoT networks to address complicated security issues in real-time which one of the attacks can freely threaten its global performance. Initially,investigate the safety performance of security intelligent prediction techniques is linking with deep learning algorithms into the IoT security risks. This contribution provides a CNN model that improves IoT security risk assessment (SRA) performance. Then, the access control techniques are changed with IoT-like dynamic systems with the number of items spread all over the place. Therefore, dynamic access control models are necessary. Thesedesign not individual use strategies of access but incorporate environmental and real-time data to predict the decision on access. The risk-based access control approach is one of those dynamic models. To decide the access decision, this model assesses the security risk value associated with the access request. This assessment of the model proposed results from the performance and accuracy of IoT networks.
Dynamic access control and security performance prediction for IoT networking using a novel deep learning technique
