NEDG: Efficient Design and Control of Heterogeneous Mobile Networks: Beyond Poisson Regime

移动性是各种应用的核心，从军事和灾难环境中搜索移动目标和救援使命的经典问题，到部署移动的ad-hoc/传感器网络用于在复杂地形和水下进行监视和数据通信。虽然这些网络中节点的随机移动模式被认为是通信链路不确定性和中断的主要来源，但如果适当控制和积极利用，它也可以促进可靠和可预测的性能。大多数现有的方法都利用节点的移动性，仅在偶然接触时进行被动的数据转发，with an over-simplifying过分简化assumption假设that every一切pair对of nodes节点are making制造contacts接触with each其他according to a Poisson泊松process过程，这与最近在各种移动的网络中经验观察到的具有高度异质性的非泊松性质相矛盾。本研究的长期目标是为异构移动ad-hoc网络和非Poisson接触下的延迟/中断容忍网络中的节点的有效协议设计和控制开发一种统一的方法。该目标进一步扩展到使用移动节点与可控的移动性，可以自主地利用thechanging多样性。这项研究的更广泛的影响在于其跨学科性质，解决了异构非泊松接触的移动性上运行的算法的分析和设计，这也是其他科学领域的核心，如流行病学，觅食生态学和生物学，所有这些都是由活生物体之间的接触驱动的。这种移动的ad-hoc网络与自然界中各种现象之间的紧密联系将为大学新生甚至普通公众提供生动的例子。

Mobility is central to various applications, from the classicalproblems of searching for a moving target and rescue mission inmilitary and disaster settings, to deploying mobile ad-hoc/sensornetworks for surveillance and data communication over hostileterrain and underwater. While the random mobility pattern of nodes inthese networks has been considered as the main source of uncertainty anddisruption of communication links, it can also facilitate reliable andpredictable performance, if properly controlled and actively exploited.Most existing approaches have taken advantage of nodes' mobility only aspassive vehicles for data forwarding upon contact by chance, with anover-simplifying assumption that every pair of nodes is making contactswith each other according to a Poisson process, which is recentlycontradicted by empirically observed non-Poisson nature with high degreeof heterogeneity in various mobile networks.The long-term goal of this research is to develop a unified methodologyfor efficient protocol design and control of nodes in heterogeneousmobile ad-hoc networks and delay/disruption-tolerant networks undernon-Poisson contacts. The goal further extends to the use of mobilenodes with controllable mobility that can autonomously exploit thechanging diversity. The broader impact of this research lies in itsinterdisciplinary nature in addressing the analysis and design ofalgorithms running on mobility with heterogeneous non-Poisson contacts,which is also central to other scientific areas such as epidemiology,foraging ecology and biology, all driven by contacts among livingorganisms. This tight relationship between mobile ad-hoc networksand various phenomena in nature will serve as a vivid example forundergraduate students at a freshmen level or even to general public.