期刊
JOURNAL OF NETWORK AND COMPUTER APPLICATIONS
卷 179, 期 -, 页码 -出版社
ACADEMIC PRESS LTD- ELSEVIER SCIENCE LTD
DOI: 10.1016/j.jnca.2021.102982
关键词
Information-centric networking; Satellite network; Fountain coding; Content delivery; Reliable networking; Delay tolerant communication; Named data networking; Coding-enabled data retrieval
类别
资金
- National Natural Science Foundation of China [92067203, 61672101]
- National Key Research and Development Program of China [2019YFB1803200, 2020YFB1806000]
The study proposes an information-centric data retrieval scheme, named Coding-enabled Adaptive-Pull (CAP), to reduce delivery delay, improve content transmission efficiency, and tolerate disruptions.
Building a reliable and efficient network layer for large-volume content delivery over satellite is extremely challenging due to space circumstances of unreliable connectivity, long propagation delay and high dynamics. In this paper, we address those challenges in satellite networking and propose an information-centric data retrieval scheme to provide disruption-tolerant, low-delay and highly efficient content delivery. Our design, namely Coding-enabled Adaptive-Pull (CAP) scheme, is rooted in the concept of information-centric networking, transferring named contents rather than maintaining unreliable end-to-end connections in space. CAP retrieves data in one-to-many request-response model and adaptively adjusts the retrieval expectation according to network states. Meanwhile, the data retrieval scheme embeds with fountain-style erasure coding over its named packets for reliable transmission to reduce the needs of packet retransmission and utilizes in-network caching to tolerate possible link disruption. Evaluation results show that our scheme reduces delivery delay in most cases of unreliable links by nearly 82% and 75% compared with IP and NDN, respectively, and by 72% compared to delay-tolerant networking technique in a space-oriented scenario, with outstanding performance in tolerating delays and disruptions.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据