Mitigation of network tampering using dynamic dispatch of mobile agents

Detection of malicious activity by insiders, people with legitimate access to resources and services, is particularly difficult in a network environment. In this paper, a novel classification of tampering modes is identified that can be undertaken by insiders against network Intrusion Detection Systems (IDSs). Five categories of tampering modes are defined as spoofing, termination, sidetracking, alteration of internal data, and selective deception. These are further distinguished specifically toward IDS sensor, control, and alarm categories such as spoonfeeding, sugarcoating, and scapegoating. The Collaborative Object Notification Framework for Insider Defense using Autonomous Network Transactions, or CONFIDANT, uses distributed mobile agents to mitigate these tampering exposures. CONFIDANT employs techniques such as encapsulation, redundancy, scrambling, and mandatory obsolescence. This paper describes how these mitigation techniques are applied within the CONFIDANT framework. The approach focuses on evaluating file integrity through the use of dynamically dispatched mobile agents.