DIMACS Workshop on Mobile and Wireless Security

November 3 - 4, 2004
DIMACS Center, CoRE Building, Rutgers University, Piscataway, NJ

Bill Arbaugh, University of Maryland, waa@cs.umd.edu
Presented under the auspices of the Special Focus on Communication Security and Information Privacy.


Nidal Aboudagga and Jean-Jacques Quisquater, UCL Crypto Group, Belgium

Title: Wireless Security and Roaming Overview

The progress and deployment of wireless network is growing very fast. It induces an increasing need for security and connectivity. At the beginning the IEEE 802.11 working group don't focus on security but tried to provide a wired equivalent privacy to WLAN via WEP protocol based on RC4 algorithm. Unfor- tunately, it failed in this task. For more access control the IEEE 802.1x is adapted to the wireless network. 802.1x combined with EAP protocol allow several kind of users authentication. To solve the problems of data security on wireless net- work, Wi-Fi alliance create the WPA protocol based on earlier works of IEEE 802.11i group. Now the IEEE 802.11i standard is approved, it use strong security features(TKIP,AES); but the challenge now is how to ensure the same level of security and uninterrupted connectivity for roaming stations.

Jari Arkko, Ericsson, Pasi Eronen, Nokia, Pekka Nikander, Ericsson, and Vesa Torvinen, Ericsson

Title: Secure and Efficient Network Access

A number attempts are currently being made to improve the efficiency, security and functionality of network access, particularly with mobile nodes. These attempts include network access authentication mechanisms, fast handover mechanisms, IP layer attachment improvements, and other optimizations focusing on individual components. In our abstract, we sketch a new architecture that focuses on the problem as a whole rather than on a single layer (link-layer) or a single function (authentication): what tasks are necessary in order to have a node attach to a network? How can that node move into another attachment point? What is the best order of the tasks so that the number of roundtrips is minimized? From a high-level point of view, an attachment to a network consists of a transaction between the mobile node, access node, router, access network, home network, possibly some mediating networks, and possibly also some mobility related nodes such as home agents. Some of these entities, such as access networks, are not explicitly addressed or identified in current designs. Our design ideas deal with the different aspects of the network access problem, have high resistance to denial-of-service attacks, and protect the privacy of the participants. Lessons from protocols such as IKEv2 and HIP have been used in the design.

Florent Bersani

Title: Network Access Control Schemes Vulnerable to Covert Channels

In this presentation, we underline a trivial yet potentially harmful vulnerability of some network access control schemes that allow an authentication server from a different administrative domain to be involved. IEEE 802.11 wireless LANs seem to resort to such vulnerable schemes in roaming scenarios. We compare the access control scheme WLANs resort to in roaming scenarios to different techniques used in the cordless or the cellular telephony realms. We then suggest different ways to mitigate its vulnerability.

Pasi Eronen, Nokia and Jari Arkko, Ericsson

Title: Role of authorization in wireless network security

Wireless security work has largely focused on authentication and key exchange, and using the resulting keys for protecting individual packets. Authorization has often been considered something that "just happens" at some step, but we think this picture is misleading. This presentation explores what implications enforcing certain kind of authorization policies has for the architecture. The focus is especially on public networks where policies that are mainly about money ("anyone who pays is allowed") are fundamentally different from policies about authenticated identities ("anyone in the RD_Employees group is allowed").

Dan Geer and Moti Yung

Title: Threshold Cryptography and Wireless Roaming

Threshold Cryptography (distributed key splitting) is traditionally employed as means to preserve the whole key against compromise, i.e. for risk reduction (memory compromise) and availability (coping with denial of service). Recently, some functionality of splitting key has been shown to be useful beyond preservation, yielding a small number of high-security, server-related applications. In this talk, we'll explore how (or how not) Threshold Cryptography may help in the roaming problem.

Scott Guthery, CTO Mobile-Mind, Inc., and Mary J. Cronin, Boston College

Title: Extending the GSM/3G Key Infrastructure

This presentation provides a brief technical overview of the GSM/3G key infrastructure and the protocols that drive it with a particular focus on the role and security architecture of the SIM. The talk then explores whether this infrastructure offers a viable model for enhancing the security of 802.11 and ad hoc mobile networks. The presenters will analyze the leading technical initiatives and business models for SIM-based security outside of GSM/3G networks today and will discuss alternative approaches to extending the GSM/3G key infrastructure into non-telephony domains.

James Kempf, DoCoMo USA Labs

Title: A Proposal for Next Generation Cellular Network Authentication and Authorization Architecture

The current wireless networks provide access authentication and authorization through two architectures:

  1. Layer 2.5 AAA flows prior to granting granting network access and Internet routing service, and upon handover between access points,
  2. HTTP Web page login that authenticates using a user name/password for account holders or a credit card number for one time access with no reauthentication on handover.

In this talk, I'll discuss an alternative architecture using certificates that could evolve from the IETF SEND router certification protocol, and, possibly, the IETF PANA protocol. I'll briefly review SEND and discuss prospects for extending it to handle network access authentication and authorization, and talk some about PANA. While this proposed architecture has some advantages, prospects for acceptance and actual deployment remain speculative at best, because of the deep embedding of traditional AAA architecture into existing wireless cellular and wireless standards and the strong hold the AAA architecture has on thinking in the wireless and cellular community.

Jesse Walker, Intel

Title: 802.11 Authentication and Keying Requirements

This talk will examine some of the issues presented by 802.11i's use of EAP. These include freshness, binding, performance, and deployability issues. The goal is to set some agenda work items for future standards development.

Previous: Program
Workshop Index
DIMACS Homepage
Contacting the Center
Document last modified on October 26, 2004.