Machine Learning based Authentication of loT Devices in Traffic Prediction for ITS

Abstract

Now a days for successful traffic modelling, management and accurate traffic flow forecast is becoming increasingly important. Traffic congestion prediction has become an ongoing research area in recent years, particularly in the field of machine learning. For years, intelligent transportation systems (ITS) have been gathering and processing massive volumes of data from a variety of sensors in order to establish a traffic ground truth. Many prediction approaches have been offered to improve traffic flow prediction performance with the Machine Learning based (ML) methods. It requires less prior knowledge of the relationships between distinct traffic patterns, which has fewer constraints on prediction tasks. To ensure a smooth flow of traffic, ITS combines machine learning with the traffic control policies and performs real-time traffic scheduling. Indeed, the next generation of IoT will necessitate a new secure-by-design approach, in which threats are handled in advance. To this aim, machine learning will be critical in providing IoT devices with both reconfigurability and intelligence. Traffic prediction can be aided by IoT combined with Artificial Intelligence, with this without the need for human involvement traffic prediction sensors connect with IoT devices collects data which requires computations in order to derive informed decisions. This paper will provide a thorough review on the machine learning techniques, which may be used to improve an application's intelligence and capabilities, as well as security concerns such as authentication, and authorization of IoT devices. To address all these concerns various approaches are developed, but most of them are dependent on expanding the computing capability, storage, and power. The machine learning concepts are used to address authentication difficulties is proposed in this paper. The proposed method is based on a machine learning model that does not rely on IoT device processing performance, storage, or power.