Real-Time Data Delivery Using Prediction Mechanism in Mobile Environments

详细信息    查看全文

• 作者：Ding-Jung Chiang (1)
  Ching-Sheng Wang (2)
  Chien-Liang Chen (2)
  Der-Jiunn Deng (3)
  
• 关键词：Kalman filter ; Prediction mechanism ; Quality of service ; Time constraint
• 刊名：Wireless Personal Communications
• 出版年：2014
• 出版时间：February 2014
• 年：2014
• 卷：74
• 期：4
• 页码：1345-1362
• 全文大小：595 KB
• 作者单位：Ding-Jung Chiang (1)
  Ching-Sheng Wang (2)
  Chien-Liang Chen (2)
  Der-Jiunn Deng (3)
  
  1. Department of Digital Multimedia Design, Taipei Chengshih University of Science and Technology, Taipei, Taiwan
  2. Department of Computer Science and Information Engineering, Aletheia University, Taipei, Taiwan
  3. Department of Computer Science and Information Engineering, National Changhua University of Education, Changhua, Taiwan
  
• ISSN：1572-834X

文摘

Mobile data delivery is a critical issue in the mobile computing area. One of the most important problems is the efficient access to data. A proposed solution to this problem is the prefetching technique which consists in putting in reserving the information before the users need it. Low bandwidth, unreliable wireless links, and frequent disconnections of mobile environments make it difficult to satisfy the timing requirements of traditional strategies. This paper investigates broadcast scheduling strategies for push-based broadcast with timing constraints in the form of deadlines ,and proposes a prediction algorithm based on Kalman filter theory for this study. The proposed dissemination policy and adaptive bandwidth allocation scheme obtain sufficient conditions such that all the time-bounded traffic sources satisfy their timing constraints to provide various quality of service guarantees in the broadcast period. Our goal is to identify scheduling algorithms for broadcast systems that ensure requests meeting their deadlines. Our approach examines the performance of traditional real-time strategies and mobile broadcasting strategies, and demonstrates that traditional real-time algorithms do not always perform the best in mobile environments. The proposed design indeed achieves good performance in mobile environments.