Cross Layer Optimization for Wireless Video Transmission Using Machine Learning

Currently demand for the applications that use video as the main content is growing fast and these multimedia applications demand a high Quality of Service (QoS). Cross-layer design uses an approach in which different layers of the architecture are cooperatively joined, and hence, delivers diverse reliability and QoS provisions for wireless multimedia networks. This in turn improves the overall performance of video transmission in real time. This paper proposes a cross layer optimization system with low computing time that selects parameters at one layer based on the parameters at other layers. The cross-layer optimization problem is resolved by using machine learning technique of classification. Various classification algorithms are employed for the cross-layer parameter selection using a generalized training dataset and the accuracy of each classifier is computed. The transmission of video encoded using the selected optimum parameters gives a better performance compared to the ad-hoc approaches.