TC: Small: Enemies from Within: Thwarting Sophisticated Insider Attacks in Wireless Networks

由于无线介质的开放性和无线设备的物理安全性差，无线网络固有地容易受到外部和内部网络攻击。虽然外部攻击可以通过基于密码的措施和鲁棒性机制的组合来中和，但从受损节点发起的内部攻击本质上要复杂得多。内部对手利用网络秘密的知识（例如，密码密钥和伪随机扩频码）和协议语义，以通过选择性地和自适应地瞄准关键网络功能来最大化它们的有害影响。该项目的重点是设计和评估安全协议，以挫败复杂的内部攻击的选择性，自适应性和跨层的性质。 考虑了统一的对抗模型，其中选择性干扰和/或丢弃高价值分组（例如，控制信息）被用作拒绝网络服务的主要工具。在这种模式下防御无线网络特别具有挑战性，因为对手在短时间内保持活跃，并且可能会在飞行中调整/修改他的策略。内部攻击对关键网络功能（如端到端可靠的数据传输和路由）的影响通过广泛的实验和建模进行量化。新的协议的开发，检测正在进行的攻击，并及时识别受损的节点。项目成果将加强无线网络对内部人员发起的拒绝服务攻击的抵御能力。 预计拟议的研究将对每一种无线技术产生深远的影响（例如，WiFi、WSN、MANN、CRN、Mesh等），由于所解决的漏洞的一般性。

Wireless networks are inherently vulnerable to external and internal network attacks, due to the open nature of the wireless medium and the poor physical security of wireless devices. While external attacks can be neutralized through a combination of cryptography-based measures and robustness mechanisms, internal attacks, which are launched from compromised nodes, are much more sophisticated in nature. Insider adversaries exploit knowledge of network secrets (e.g., cryptographic keys and pseudo-random spreading codes) and protocol semantics to maximize their detrimental impact by selectively and adaptively targeting critical network functionalities. This project focuses on designing and evaluating secure protocols for thwarting sophisticated insider attacks of selective, adaptive, and cross-layer nature. A unified adversarial model is considered in which selective jamming and/or dropping of high-value packets (e.g., control information) is used as the primary vehicle for denying network service. Defending wireless networks under this model is particularly challenging, because the adversary remains active for short periods of time and may adapt/modify his strategies on the fly. The impact of insider attacks on critical network functions such as end-to-end reliable data delivery and routing is quantified via extensive experimentation and modeling. Novel protocols are developed which detect ongoing attacks and identify compromised nodes in a timely manner. Project results will reinforce the resilience of wireless networks to denial-of-service attacks launched by insiders. The proposed research is expected to have profound impacts on every wireless technology (e.g., WiFi, WSNs, MANETs, CRNs, Mesh, etc.), due to the generality of the addressed vulnerabilities.