首页> 外文期刊>IEEE transactions on wireless communications >Inter-Session Network Coding-Based Policies for Delay Tolerant Mobile Social Networks
【24h】

Inter-Session Network Coding-Based Policies for Delay Tolerant Mobile Social Networks

机译:基于会话间网络编码的容错移动社交网络策略

获取原文
获取原文并翻译 | 示例
           

摘要

We consider delay tolerant mobile social networks (DTMSNs), which are opportunistic networks made of human-carried wireless devices clustered into social communities. In such environments, routing is a challenge as the limited resources (such as memory and contact opportunities) must be efficiently used and shared between the sessions (or users, contents). To handle several unicast sessions, inter-session network coding (ISNC) has been proved necessary for optimal throughput in general networks, but is a delicate problem as it can quickly get detrimental. This paper investigates that ISNC can be beneficial to DTMSNs when used on top of a social-aware routing algorithm, whereas we exemplify and make explicit why any gain can hardly be expected with greedy replication, in regard to the current literature on ISNC. We then design decentralized criteria to control when and where in the network ISNC should be triggered, based on the node features (buffer size and social relationships) and network current load. These criteria are tested extensively on real-world contact traces, in terms of various metrics, such as number of deliveries, mean delay, or fairness. Our online ISNC protocol builds on the SimBet utility-routing policy. Our ISNC protocol can, however, run on top of any social-aware routing.
机译:我们考虑耐延迟的移动社交网络(DTMSN),这是由人类携带的无线设备组成的机会网络,这些无线设备聚集在社交社区中。在这样的环境中,路由是一个挑战,因为必须有效使用有限的资源(例如内存和联系机会)并在会话(或用户,内容)之间共享。为了处理多个单播会话,会话间网络编码(ISNC)已被证明对于在常规网络中实现最佳吞吐量至关重要,但是由于它会很快变得有害,因此这是一个棘手的问题。本文研究了在社交感知路由算法之上使用ISNC可能对DTMSN有益,而根据当前关于ISNC的文献,我们举例说明了为什么用贪婪复制很难获得任何收益。然后,我们根据节点功能(缓冲区大小和社会关系)和网络当前负载,设计分散的标准,以控制何时应在网络中触发ISNC。这些标准已根据各种指标(例如交货数量,平均延误或公平性)在现实世界中的联系轨迹上进行了广泛测试。我们的在线ISNC协议基于SimBet实用程序路由策略。但是,我们的ISNC协议可以在任何具有社交意识的路由上运行。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号