Reduction of Packet Error Rate in V2V Communication Based on Machine Learning

Reducing the accidents that resulted from traffic is a highly concerning issue. Automating cars or emergency communication are attempting the decrease these accidents. So improving the conditions of communication between vehicles under research. One of the improvements comes by standardizing transmitting packets, which is best executed by the 802.11p protocol. 802.11p encompasses a field that has the size value of the transmitted packet.. So this work centering on developing a powerfully adaptable packet size framework, which depends on the value of the signal-to-noise ratio (SNR). There is neural network (NN) controls to the framework, which prepared by candidate values of packet size come from an equation derived practically by testing it numerous times the connection between packet error ratios (PERs) and packet size values. The usage of the proposed framework comes about in a noteworthy diminishment in PER, and the comes about were appeared by practice the proposed framework and two other frameworks, one of them is satisfied without channel tracking whereas the moment had channel tracking. The comparison for band 10MHz appears there's a 46.6 percent and 36.88 percent average decrease in PER values compared to not tracking the channel and tracking the channel, separately. Comes about average improving the PER values for band 20MHz were 18 percent for the without-channel-tracking and 66.91 percent for the with-channel-tracking frameworks. As a result, the inspiration to utilize NN, which is one field of machine learning (ML), was figured out. Besides, this work persuades more researches that got to be done applying the ML into the vehicle to vehicle (V2V) communication.