Research program in computer and network security

该提案描述了达尔豪西大学将在新成立的隐私和安全实验室的主持下开展的计算机和网络安全研究方案。拟议的计划包括观察和实验研究。多年来，互联网已经发展到表现出紧急行为的地步，即无法从组成互联网的单个系统或单个系统的分组的行为中预测到的聚合现象。这些现象以各种方式表现出来，包括在扫描蠕虫爆发期间核心路由结构的部分崩溃，以及相对较小的感染Welchia-B蠕虫的机器池对其他幂定律连接统计数据的高度明显干扰。通过聚合和分析大量网络数据，我们培养对网络的全球情景意识，使我们能够正确地看待当地的观察。同时，单个企业和系统的防御需要本地检测和防御措施。*已经有几次尝试对入侵检测系统等防御机制进行评估，但由于多种原因，这些评估在很大程度上并不令人满意。*拟议研究计划的第二个主要推动力是创建一个隔离的测试设施，可用于各种涉及恶意代码的项目，如蠕虫，并探索检测和缓解对终端系统和网络的攻击的新方法。这一测试设施还将支持多个相关领域的研究项目，包括生成用于安全测试的真实背景流量，以及评估入侵检测和防御系统的更好技术。此外，它还可以作为专用的模拟或仿真设施，以探索更大网络的网络行为的选定方面。该计划还为研究人员和从业人员提供了计算机安全实践方面的培训机会。

This proposal describes a program of research in Computer and Network security to be carried out at Dalhousie University under the auspices of the newly established Privacy and Security Laboratory. The proposed program covers both observational and experimental studies.  Over the years, the internet has grown to the point that it exhibits emergent behaviors, i.e. aggregate phenomena that cannot be predicted from the behaviors of the individual systems or groupings of individual systems from which it is composed.  These manifest themselves in a variety of ways, including the collapse of parts of the core routing structure during scanning worm outbreaks, and a highly visible disturbance of an otherwise power law connection statistic by a relatively small pool of machines infected with the Welchia-B worm.  By aggregating and analyzing massive amounts of network data, we develop global situational awareness of the network that allows us to place local observations in perspective. At the same time, the defense of individual enterprises and systems requires local detection and defense measures.  There have been a few attempts at performing evaluations of defensive mechanisms such as Intrusion Detection Systems, but these have been largely unsatisfactory for a number of reasons.  The second major thrust of the proposed research plan is the creation of an isolated test facility that can be used for a variety of projects involving malicious code such as worms and for exploring new ways to both detect and mitigate attacks on end systems and networks. This test facility will also support research projects in a number of related areas, including the generation of realistic background traffic for use in security testing, and better techniques for evaluating intrusion detection and prevention systems.  It can also serve as a dedicated simulation or emulation facility to explore selected aspects of network behavior for larger networks. The program also provides opportunities for the training of both researchers and practitioners in computer security practices.