期刊名称：International Journal of Electrical and Computer Engineering
电子版ISSN：2088-8708
出版年度：2016
卷号：6
期号：6
页码：3205-3216
DOI：10.11591/ijece.v6i6.pp3205-3216
语种：English
出版社：Institute of Advanced Engineering and Science (IAES)
摘要：Although there has been a tremendous rise in places being connected through the Internet or any other network protocol, there still lie areas, which remain out of reach due to various reasons. For all such places the answer is a Delay Tolerant Network (DTN). A DTN is such a network where there is no fixed or predefined route for messages and no such guarantee whatsoever of all messages being correctly routed. DTN can be considered as a superset of networks wherein other networks such as adhoc, mobile, vehicular etc. form the subset. Therefore routing in DTN is a very chancy affair where one has to maximize on the present network scenarios to get any fruitful result other than depending on past information. Also protocols here need to be less complex and not increase the already high nodal overhead. In this paper we propose a new approach, the Fibonary Spray and Wait, which does exactly this. It forwards copies of a message in a modified Binary Spray and Wait manner so that it performs well even in non independent and identically distributed node structure. We have supported our statements with mathematical as well as simulation analysis.
其他摘要：Although there has been a tremendous rise in places being connected through the Internet or any other network protocol, there still lie areas, which remain out of reach due to various reasons. For all such places the answer is a Delay Tolerant Network (DTN). A DTN is such a network where there is no fixed or predefined route for messages and no such guarantee whatsoever of all messages being correctly routed. DTN can be considered as a superset of networks wherein other networks such as adhoc, mobile, vehicular etc. form the subset. Therefore routing in DTN is a very chancy affair where one has to maximize on the present network scenarios to get any fruitful result other than depending on past information. Also protocols here need to be less complex and not increase the already high nodal overhead. In this paper we propose a new approach, the Fibonary Spray and Wait, which does exactly this. It forwards copies of a message in a modified Binary Spray and Wait manner so that it performs well even in non independent and identically distributed node structure. We have supported our statements with mathematical as well as simulation analysis.
