A Cognitive Policy Framework for Next-Generation Distributed Federated Systems: Concepts and Research Directions
A Cognitive Policy Framework for Next-Generation Distributed Federated Systems: Concepts and Research Directions
复制标题
下一代分布式联邦系统的认知政策框架：概念和研究方向
DOI：
--
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
B. Rivera
中科院分区：
文献类型：
--
作者：
E. Bertino;S. Calo;Maroun Touma;D. Verma;Christopher Williams;B. Rivera
下一代协作活动和任务将由具有各种认知能力的自主设备组执行。这些设备必须在不确定、不安全（物理和网络）和不稳定的环境中运行。在这样的环境中，通信可能是分散的。因此，对这种自主设备组进行基于策略的适当管理至关重要。然而，目前的政策管理系统有许多局限性，包括缺乏灵活性。在本文中，我们阐述了新的架构方法解决的要求，有效管理的自治组的设备，并讨论了生成政策的概念-一种新的范式，提高了灵活性的基于政策的方法来管理。在本文中，我们还调查了管理设备组所必需的策略类型。尽管许多这样的政策类型存在于传统的环境中，他们在我们的背景下使用提出了新的挑战，我们阐明的文件。我们还介绍了一个研究路线图，讨论了几个研究方向的发展认知和灵活的基于政策的方法来管理自主组的设备协作任务。最后，由于我们提出的政策范式是数据密集型的，我们讨论的问题，提供所需的数据，在环境中的政策决策的特点是流动性，不确定性，和分散的通信。
Next-generation collaborative activities and missions will be carried out by autonomous groups of devices with a large variety of cognitive capabilities. These devices will have to operate in environments characterized by uncertainty, insecurity (both physical and cyber), and instability. In such environments, communications may be fragmented. Proper policy-based management of such autonomous device groups is thus critical. However current policy management systems have many limitations, including lack of flexibility. In this paper, we articulate novel architectural approaches addressing the requirements for the effective management of autonomous groups of devices and discuss the notion of generative policies - a novel paradigm that enhances the flexibility of policy-based approaches to management. In this paper, we also survey types of policy that are essential for managing device groups. Even though many such policy types exist in conventional settings, their use in our context poses novel challenges that we articulate in the paper. We also introduce a research roadmap discussing several research directions towards the development of a cognitive and flexible policy-based approach to the management of autonomous groups of devices for collaborative missions. Finally, as our proposed policy paradigm is data-intensive, we discuss the problem of supplying the data required for policy decisions in environments characterized by mobility, uncertainly, and fragmented communications.