M.A. Thesis:

A new traffic control system for IoT networks using neural fuzzy algorithm and locust mutation

With the emergence of the Internet of Things, one of the most important fields that benefited from this phenomenon is the field of transportation and traffic. From the smart parts used in cars to the sensors and equipment installed on the roads, all by connecting to the Internet, it has made it possible to manage and operate them more usefully and easily. Traffic control is one of the important issues in the field of transportation, which can be significantly improved with the help of the Internet of Things. The Internet of Things, like many technologies, in addition to the advantages they bring with them, also have problems and disadvantages. Among the problems that the Internet of Things brings is the explosive volume of information produced by various objects that must be solved. The amount of information input to the system can be so large that it causes the analysis to deviate from the actual result and also slow and cost the calculations. Therefore, in this thesis, in order to reduce the dimensions and select the most suitable features that can represent the input data, the grasshopper algorithm has been used to reduce the dimension of the features. In order to control the traffic, several methods have been used so far, and in many methods, different types of neural networks have been used. In this thesis, ANFIS neural fuzzy algorithm is used to train the system and predict traffic. As will be seen at the end, the proposed method will significantly improve performance and reduce errors.

Keywords: system, traffic, neural, design, internet, problems, reduction