Nowadays, with the growth of urbanization and the increase in the number of cars in many developing countries, it has increased the density and the number of traffic accidents in the urban road network, which has not been designed to handle this volume and type of traffic. In fact, traffic is one of the new challenges of humanity in large and densely populated cities, and although some solutions have been devised to solve it, it seems that many of them have been unsuccessful so far. Therefore, in this research, a new approach to optimizing the traffic control of the intelligent transportation system based on data mining and evolutionary algorithms is presented. The proposed method is a data recognition scheme based on edge computing for intersection traffic lights, where the traffic lights act as edge nodes for recognizing vehicle data. In the proposed method, analytical results are generated on the controllability, stability, and accessibility of a mixed traffic system consisting of data related to it on a road with a single or multiple intersections. The working method is as follows: a single intersection scenario is considered, and with the help of V2E communication, vehicle data is collected from the base station and the Quoshen filter is used to verify the reliability and accuracy of the data. Then, a multiple intersection scenario is considered, and the vehicle data of two adjacent intersections is merged, and then the reliability of the data is verified by the Quoshen filter. Another point is that this scheme uses the mmh3 hash functions available in the QF Quoshen filter to reduce the occupied space of the edge node computational resources and the bit error rate. Since the proposed scheme results in the reliability and effectiveness of vehicles with less delay, it can be said that the use of the data recognition approach is effective in quickly recognizing data even with a large number of vehicles and their complex data. By using the combination of the Quoshen filter and the particle swarm evolutionary algorithm in the proposed method, an improvement of 20 milliseconds has been recorded in the results.
Samadi,M R and . khavasi,A . (2023). A new model to optimizing traffic control in intelligent transportation systems based on data mining and particle swarm evolutionary algorithm. Computing and distributed systems, 6(1), 89-103.
MLA
Samadi,M R , and . khavasi,A . "A new model to optimizing traffic control in intelligent transportation systems based on data mining and particle swarm evolutionary algorithm", Computing and distributed systems, 6, 1, 2023, 89-103.
HARVARD
Samadi M R, . khavasi A. (2023). 'A new model to optimizing traffic control in intelligent transportation systems based on data mining and particle swarm evolutionary algorithm', Computing and distributed systems, 6(1), pp. 89-103.
CHICAGO
M R Samadi and A . khavasi, "A new model to optimizing traffic control in intelligent transportation systems based on data mining and particle swarm evolutionary algorithm," Computing and distributed systems, 6 1 (2023): 89-103,
VANCOUVER
Samadi M R, . khavasi A. A new model to optimizing traffic control in intelligent transportation systems based on data mining and particle swarm evolutionary algorithm. Computing and distributed systems. 2023;6(1):89-103 (In Persian).