NeTS-NBD: Dynamic Carrier-Assisted Routing in Mobile Networks

移动网络中的载波辅助路由在处理各种移动网络中的节点移动性方面，存在两种学派。传统的基于连接的模型是建立在底层网络是连通的前提下的，认为节点的移动性是不可取的。移动性辅助模型认为移动性是一个理想的特征，其中路由是基于存储-向前范式的随机或控制移动节点的移动。本研究利用节点移动性来提升移动网路的路由能力。提出了一种双控制平面模型，其中包括传统的S（静止）平面，用于在静止节点之间使用基于连接的消息路由。另外，提出了一种新的控制平面M（移动）平面，用于移动节点的轨迹控制。重点是移动节点的松散控制运动，以协助路由。本工作的五个目标如下：(1)提出一个通用的路由框架。(2)提供灵活的移动节点轨迹控制，平衡多个全局目标。(3)探索使用单播作为支持集体通信的基本构建块，包括组播、广播、任播、聚集和八卦。(4)将目前二维欧几里得空间中的模型扩展到高维非欧几里得空间。(5)在双控制平面中集成不同的组件，并基于一组定义良好的定量性能指标通过实证研究对系统进行微调。这项工作首次集成了处理节点移动性的两种学派。PI设想，从这项研究中产生的结果将为包括GENI项目在内的广泛应用中的有效路由提供指导。

Carrier-Assisted Routing in Mobile NetworksIn terms of handling node mobility in various mobile networks, two schools exist. The traditional connection-based model is built on the premise that the underlying network is connected, and views node mobility as undesirable. The mobility-assisted model considers mobility as a desirable feature, where routing is based on the store-carry-forward paradigm with random or controlled movement of mobile nodes. This research uses node mobility to enhance routing capability in a mobile network. A dual control planes model is presented, which includes the traditional S (stationary)-plane for routing among stationary nodes using connection-based message routing. In addition, a new control plane, M (mobile)-plane, is used for trajectory control of mobile nodes. The focus is on loosely controlled movement of mobile nodes to assist routing. The five goals of this work are the following: (1) Propose a generic framework for routing. (2) Offer flexible trajectory control of mobile nodes to balance several global objectives. (3) Explore the use of unicasting as a basic building block to support collective communication, including multicast, broadcast, anycast, gather, and gossip. (4) Extend the current model in the 2-D Euclidian space to the high-dimensional and non-Euclidian space. (5) Integrate different components in dual control planes and fine tune the system through an empirical study based on a set of well-defined quantitative performancemetrics. This work is the first to integrate two schools of handling node mobility. The PI envisions that results generated from this research will provide guidelines for efficient routing in a wide range of applications, including the GENI project.