SGER: A Lightweight Drive-Through Management Service for Internet-enabled Ad-Hoc Networks

点击翻译按钮获取中文摘要

As the Internet evolves into a global communication infrastructure, it is becoming complex and ubiquitous. It is bringing together heterogeneous collections of hosts and network devices such as wireless mesh networks. A consequence of this ever-increasing scale, complexity and agility is that problems of managing this infrastructure keep getting worse. Research concentrating on streamlining fundamental issues has shown that absence of a single end-to-end lightweight hybrid network management solution with the necessary levels of context awareness and adaptability, while much desirable, has not yet been realized. Consequently, the community is still detained from fully gaining from many of the emerging applications of the cyber-infrastructure. This research involves an exploratory investigation towards a ?drive through? network management platform for ?on-the-fly? Internet-enabled mobile networks. Objectives include new conceptual approaches that couple application, and middleware designs with design of network management services; novel theoretical models that tap into in-network management characteristics of pervasive computing so that missing or conflicting information gathered from multiple participating nodes can be tolerated; augmentation of security management algorithms with trustworthiness functionalities; theoretical schemes that examine epistemic uncertainty by fusing imprecision, uncertainty and event synchronization with belief theory; and extensions to archetype platforms for supporting autonomic network management and configuration. The intellectual merit of the research lies in presenting a first step towards synthesis of integrated network management theoretical and design solutions that support traditional network management protocols and functionalities in addition to the dynamic aspects of infrastructures-less networks. Although some risk exists, focused research can overcome the current barriers. The broader impact of the research is that this work will pave the way to transformative next generation network management paradigms and tools.