NETWORK CONGESTION PREDICTION USING MACHINE LEARNING APPROACH
Loading...
Files
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Malaysia Sarawak (UNIMAS)
Abstract
Description
The quality of service (QoS) is one important feature of the networks to guarantee the satisfaction of users and efficiency in their operations. As the number of users keep increasing, the burden on the network increase. The original network needs to handle more packet that it can. This may result in happening network congestion with the effect of the performance degradation and a decline in service quality. Network congestion is still a major problem in large-scale networks even it can solve by simply add the bandwidth but is not the best way. In order to solve this problem, a machine learning-based network congestion prediction system has been designed. This project will also develop an extension to a network monitoring system that enable it to predict the network congestion as the way to plug in the Machine Learning. This system will forecast the trends in bandwidth utilization and delay to identify congestion occurs, preventive actions can be taken in advance. As a cutting-edge tool for IT teams, the suggested solution will allow proactive network management and minimise the negative effects of congestion by accurately predicting possible congestion occurrences. Finally, the proposed system can also contribute to cost savings by optimizing network usage, improving user experience, and guiding infrastructure investments. Moreover, investment in network devices and network planning can be better allocated.