QoS intelligent prediction for mobile video networks: a GR approach
- PDF / 1,638,967 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 71 Downloads / 199 Views
(0123456789().,-volV)(0123456789(). ,- volV)
S.I. : SPIOT 2020
QoS intelligent prediction for mobile video networks: a GR approach Lingwei Xu1,2,3 • Han Wang4 • Hui Li1
•
Wenzhong Lin2 • T. Aaron Gulliver5
Received: 12 July 2020 / Accepted: 10 October 2020 Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract With the growth of mobile devices, consumer networks make the life more convenient and faster. Consumer networks consider mobile video as an important communication mode. Mobile video transmission faces complex environments, and the quality of service (QoS) of mobile video networks is very important for mobile entertainment applications. To evaluate the QoS of mobile video networks, outage probability (OP) is an important criterion. However, the mobile video networks gradually become complex, dynamic, and variable, which make it increasingly more difficult to predict the OP performance. In this paper, we investigate the OP performance analysis and prediction. The OP expressions are derived in exact closed-form. Then, based on the characteristics of mobile data, we have established a prediction model based on generalized regression (GR) neural network. A GR-based OP performance intelligent prediction algorithm is proposed. Compared with other methods, our proposed approach can obtain a better prediction effect. The prediction accuracy of the proposed approach can be increased by 64% and 58%, respectively. The running time is also the shortest. Keywords Mobile video networks Quality of service Performance analysis Performance prediction
1 Introduction In recent years, with the rise of the consumer economy, mobile devices dramatically increase in consumer groups. Consumer networks take the consumer as the service center, aiming at the individual user to enhance the consumer experience, so that life becomes more convenient and faster [1]. The essence of consumer networks is personal virtualization, which can enhance personal life & Hui Li [email protected] 1
College of Information Science and Technology, Qingdao University of Science and Technology, Qingdao 266061, China
2
Fujian Provincial Key Laboratory of Information Processing and Intelligent Control (Minjiang University), Fuzhou 350108, China
3
Key Laboratory of Opto-technology and Intelligent Control, Ministry of Education, Lanzhou Jiaotong University, Lanzhou 730070, China
4
Institute of Data Science, City University of Macau, Macau 999078, China
5
Department of Electrical and Computer Engineering, University of Victoria, Victoria, BC V8W 2Y2, Canada
consumption experience. Compared with the traditional consumption patterns, the mobile consumer applications increase the flexibility. However, these applications also produce significant amounts of data. The big data require the development of communication infrastructure, especially the wireless transmission technology [2]. To support entertainment big data transmission, fifthgeneration (5G) mobile communications have been widely employed in consumer networks [3–5]. 5G mo
Data Loading...
