PROJECT TITLE :

Network Traffic Prediction Model Considering Road Traffic Parameters Using Artificial Intelligence Methods in VANET

ABSTRACT:

Establishing Vehicular Ad hoc Networks, or VANETs, on intelligent vehicles that are capable of having Vehicle-to-Vehicle (V2V) and Vehicle-to-Road Side Units (V2R) Communications allows for the creation of these networks. In this paper, we propose a model for predicting network traffic by taking into consideration the parameters that can lead to road traffic occurring. Specifically, this model takes into account the number of users on the network. The proposed model incorporates an algorithm known as Random Forest- Gated Recurrent Unit- Network Traffic Prediction (RF-GRU-NTP) in order to make predictions about the flow of network traffic based on the traffic that is occurring simultaneously on roads and networks. This model consists of three stages, the first of which involves predicting network traffic based on V2R Communication, the second of which involves predicting road traffic based on V2V communication, and the third stage involves predicting network traffic while taking into account road traffic based on V2V and V2R communication. The hybrid proposed model, which is implemented in the third phase, uses the Random Forest (RF) Machine Learning algorithm to select the important features from the combined dataset (which includes V2V and V2R Communications). After that, the Deep Learning algorithms are applied to predict the network traffic flow, and the Gated Recurrent Unit (GRU) algorithm gives the best results. According to the findings of the simulations, the RF-GRU-NTP model that was proposed has superior performance in terms of both execution time and prediction errors when compared to other algorithms that are used for network traffic prediction.