Congestion control in network using machine and deep learning

Abstract

The Internet of Things (IoT) is a rising infrastructure of the 21st newlinecentury. The classification of traffic over IoT networks has attained newlinesignificant importance due to rapid growth of users and devices. It is the need newlineof the hour to isolate the normal traffic from the malicious traffic and to newlineassign the normal traffic to the proper destination. So as to suffice the Quality newlineof Service (QoS) requirements of the IoT users along with traffic predictions. newlineDetection of malicious traffic can be done by continuously monitoring traffic newlinefor suspicious links, files, connections created or received, unrecognized newlineprotocol/port numbers, and suspicious Destination/Source IP combinations. newlineA proficient classification mechanism in IoT environment should be efficient newlineenough to classify the heavy traffic swiftly, to deflect the malevolent traffic newlineon time and to transmit the benign traffic to the designated nodes for serving newlinethe needs of the users. Prediction of network traffic is a very important newlinefunction of any network. Traffic prediction is important to ensure good newlinesystem efficiency and ensure service quality of IoT applications, as it relies newlineprimarily on congestion management, admission control, allocation of newlinebandwidth to the system and the identification of anomalies. newlineVI newlineIn the first stage of research newline