A talk at the University of Massachusetts in Lowell, MA tomorrow (Wednesday): -------- Original Message -------- Subject: [Msgs] Reminder--Computer Science Colloquium, 2:45-4:00, October 15 Date: Tue, 14 Oct 2003 17:07:18 -0400 From: Gary Livingston To: cypherpunks@lne.com X-Orig-To: CC: Gary Livingston Colloquium Announcement Department of Computer Science UMass Lowell Title: Anonymous and Untraceable Communication in Mobile Wireless Networks Speaker: Jiejun Kong Time & Place: October 15th (Wednesday), 3:00-4:00, 311 Olsen Hall; snacks from 2:45-3:00 Abstract: Privacy in mobile wireless networks has different semantics from the traditional notion for banking systems and the wired Internet. In addition to traditional content privacy, mobile privacy also addresses security concerns for mobile node's identity and location, namely anonymity and location privacy. In this talk I will discuss anonymity and location privacy attacks as well as their countermeasures in mobile ad hoc networks, which can instantly establish a communication structure for civilian and military applications. We focus on passive routing attacks in hostile environments like battlefront. Anonymity and location privacy issues are critical for such scenarios, as allowing adversaries to trace network routes and infer the motion pattern of nodes at the end of those routes may pose serious threats to covert operations. The highly raised privacy demand poses challenging constraints on routing and data forwarding. ANonymous On Demand Routing (ANODR) is a multi-hop on demand routing scheme that can prevent wireless adversaries from compromising a mobile ad hoc network's anonymity and location privacy. ANODR provides anonymity service by dissociating the routing scheme from any naming scheme of network member's identity/address. This approach immediately achieves location privacy, and differentiates ANODR from other ad hoc routing schemes that mainly rely on nodes' address in data forwarding. It is verified by our simulation that the performance of (anonymous-only) ANODR is comparable to common on demand routing schemes currently in use (e.g., AODV). In addition, ANODR also implements untraceable routes so that passive adversaries cannot trace a packet flow to its source and sink. ANODR pays reasonable cost, such as neighborhood traffic mixing, to meet this privacy demand. It is verified by our simulation that the performance of (anonymous+untraceable) ANODR is more efficient than its peers designed for wired networks (e.g., MIX-Net). The underlying anonymity model of ANODR is a new one based on wireless broadcast, a ready-made mechanism in wireless networks and on-demand routing discovery processes. Though related research (Shields & Levin, CCS'00) has shown wired IP multicast can help anonymity, wireless broadcast is never used to achieve anonymity before ANODR's proposal "broadcast with anonymous trapdoor assignment". It is expected that our future work along this direction will lead to new means to provide anonymity and untraceability services to mobile wireless networks. Biographical Sketch of the Author: Jiejun Kong (jkong@cs.ucla.edu) is currently a Ph.D. candidate in Computer Science Department, University of California at Los Angeles (UCLA). He is interested in designing efficient, scalable, and robust security solutions for mobile wireless networks. His research topics include providing authentication/authorization/access control (AAA), secure routing, intrusion detection, and mobile privacy services to mobile ad hoc networks, in particular those with challenging network constraints and with very high security demands. Recently he focuses on anonymous and untraceable routing schemes. He has contributed to the design, implementation, and testing of network security protocols within ONR MINUTEMAN project, STTR project, and NSF iMASH project. _______________________________________________ msgs mailing list msgs@weblab.cs.uml.edu http://weblab.cs.uml.edu/mailman/listinfo/msgs -- Howie Goodell Howie@GoodL.org http://goodl.org Hardware control Visualization User interface UMassLowell Computer Science Doctoral Candidate