Techniques and Tools for Supporting
Secure Information Sharing and Collaborative Work
A proposal for NISSC Grants related to Homeland Security and Homeland Defense for
2004-2005 Period (Summer 2004, Fall 2004 and Spring 2005)
By C. Edward Chow
1. Introduction:
The objective of the proposed project is to investigate techniques and develop tools for supporting secure information sharing and collaborative work among multiple agencies. Procedures for enforcing and tracking secure information distribution will be investigated. Public key and attribute certificate management techniques will be enhanced to address the scalability and access control issues. Scalable web-based information sharing and collaborative work techniques will be developed utilizing existing infrastructure provided by content delivery networks. We will involve NISSC and USNorthcom in the planning and evaluation of the project and submit joint proposals to DoD and NSF.
Secure information sharing among multiple agencies not only needs to provide basic support for classifying information, encrypting data, ensure interoperability, but also address the controlling access to resources to multiple agencies-agency access to resources and adapting to different user profiles and interface. Information sharing and security in dynamic coalitions is a complex task, which manifests itself throughout the lifetime of the coalition [1,2]. Phillips et al [1] identifies that the critical issues that arise during a coalition’s formation, and in support of its day-to-day management and usage, includinge, but are not limited to the following:
- Federate groups of users quickly and dynamically in response to a crisis.
- Bring together resources (e.g., COTS, databases, legacy systems, etc.) without modification for usage in support of the crisis.
- Dynamically realize and manage a security policy during simultaneous crises.
- Identify users by their roles to finely tune their access in support of a crisis.
- Authorize, authenticate, and enforce a scalable security policy that can be managed and changed in response to the needs of the coalition.
- Provide a distributed security solution in support of dynamic coalitions that is portable, extensible, and redundant for survivability.
- Offer robust security policy definition, management, and introspection capabilities that are able to track and monitor system behavior and activities of users.
Even if the number of agencies involved is fixed, the participants from those agencies can be changed from tasks to tasks. Therefore the inherent technical challenges do not diminish in any way.
It was identified that the discretionary and mandatory access control mechanisms do not scale well in large scalelarge-scale multi-domain or multi-agency information sharing. Role-based access control provides more flexibility and the ability to change dynamicallychanges [10]. Recent access control research in grid computing [24,25] has proposed the use of attribute certificate for large scalelarge-scale geographical diverse collaboration, which binds the users with access control of specific resources [16, 20]. For tracking the usage of the documents, digital copyright products [21,22] and related access right languages were developed [23].
To allow multiple agencies to cooperate on urgent critical tasks, we need both secure information sharing and collaborative tools. Computer Supported Collaborative Work (CSCW) has been extensively studied [5-9]. Recent research has been focused on scalable multicast and group security support [5,12-14, 26].
Web-based techniques and tools have been demonstrated to be easy to use and reliable. The content delivery networks (CDN), such those provided by Akamai [3], have improved web access performance. CDN service is currently limited to caching of documents. Edge cache server side include (ESI) is a de facto standard intends to provide dynamic composition of web pages and to improve better web access time. The main limitations of infrastructure based CDNs like Akamai is their cost and uncertainty about how such a CDN would behave when faced with flash crowd traffic at many or all of the sites hosted by the CDN [17]. This calls for cooperation among various Edge servers. New ideas such as secure collective internetInternet defense [15] was proposed to utilize proxy servers, such as those the thousand of servers provided in CDN, as connection relay servers for enhancing the security and performance of Internet connections. Edge Server Side Include for Collaboration (ESIC) was proposed to extend ESI language for supporting the collaborative applications in as much transparent a manner as possible, and leverage on the existing infrastructure and processing power provided by the CDN [11]
2. Proposed Work
We proposed to carry out the following research tasks for supporting secure information sharing and collaborative work:
- Key and attributed certificate management for large scalelarge-scale information sharing and collaborative work.
Develop scalable efficient key and attribute certificate management techniques with multiple domains through simulation study and prototyping effort. Extend existing attribute certificate software package from Akenti project [16] for the secure information sharinginformation-sharing prototype. - Infrastructure support for secure web-based collaborative applications.
Develop network/system technologies for supporting large scalelarge-scale web-based collaboration. We shall also look into exising Peer-To-Peer collobaration techniques like Guntella [31]. In Merlin Vincent’s ESI for Collaboration Project [11], it was shown that exposing the proxy server IP address to a web browser will trigger a security alert and disallow access and packet exchange through proxy servers. One solution is to use a shared virtual IP address among the set of proxy servers and original web servers [30]. In traditional virtual clusters, all front endfront-end dispatchers are located in the same local site. How toThe deployment of wide area virtual clusters with the proxy servers contributed by participating agencies is a challenging research problem. Here it will be difficult to share a single virtual IP address with a collection of proxy server configured with multiple physical IP addresses from multiple domains. We will propose feasible solutions to this problem. The ESIC-based Java applications developed by Merlin will be ported as Java applets. We will investigate how new clustering concepts can be extended and to work with such Java applets. We will also investigate the performance impact with the use of the load balancing technique and the number of proxy servers through simulation. - Ubiquitous computing for sharing sensor and web information.
Enhance the Keystone group rekeying system [12] with secure reliable multicast [14] for supporting group communications. Investigate how mobile ad hoc users access and share sensor network information, and retrieve remote web server information. We are interested in investigating how the system adapts to users with different profiles and interfaces. Investigate how agent-based ubiquitous computing technology, such as those reported in [27, 28], supports notification and information sharing. Study how efficient group key management systems can be used to enhance the security of hybrid sensor/mobile ad hoc networks. This will be based on the software we have developed in the Secure Groupware for First Responders [13] and First Responder Sensor Networks [29] projects.
We will work with NISSC and USNorthcom USNorthCom staff, to fine tune the above tasks and submit joint proposals to DoD or NSF.
3. Answers for Questions in Proposal Requirements
a) What homeland security/defense-related topic(s) will be investigated?
We will address the following topics raised by USNORTHCOM research questions:
5. “Given terrorist access to open source information, how will the use of the Internet by non-military agencies to exchange information, plans, and procedures impact Homeland Security and/or Homeland Defense?”
By sharing secure information sharing technologies or providing easy to use systems with document tracking capability for non-military agencies to exchange sensitive information, plans, and procedure, it will help enhance Homeland Security and Homeland Defense.
11. “What are the “state of the art” methods to integrate intelligence threat information using Geographic Information Systems (GIS) and the Global Command and Control System (GCCS)?”
The proposed ESIC-based infrastructure support for collaborative work and scalable key/attribute certificate can improve performance and assist the integration of intelligence threat information using GIS and GCCS.
13. How can we best encourage federal, state and local governments to participate more fully in appropriate exercises?
The easy to use web-based tools and scalable predictable performance developed for secure information sharing and collaborative work will encourage federal, state, and local governments to participate more fully in appropriate exercises.
20. What is the ideal C2 construct for service components in support of combatant commander’s missions?
The techniques and tools developed in Ubiquitous ubiquitous computing for sharing sensor and web information and ESIC-based infrastructure supports partially address partially the above questions.
26. What are the most effective vehicles for establishment of a Homeland Defense/Homeland Security repository of reference material/documents and lessons learned?
The proposed secure ubiquitous notification with CDN’s proven caching/distribution capability should provide viable effective vehicles for such a mission.
b) What research capability currently exists related to the proposed work?
In the Computer Science Network and Systems Lab, we have developed Linux-based content switches (LCS and LACS) for routing packets based on IP/Port/SSL Session ID/URL/XML content of the incoming packets and a set of switching rules. We have built an 8 node distributed network using Fedora Core 1 Linux routers. It can be used for prototyping and testing the software for the proposed work. We have developed version 1 of secure groupware for first responders (SGFR). Keystone group key management software was modified and integrated with the Jabber instant messaging system. SGFR also supports remote group file distribution and remote display. It was ported to run on IPAQ PDA with NIST mobile ad hoc network driver. We have purchased and developed TinyOS-based programs with Crossbow mica2 mote sensors. For details about the equipments, please see
Dr. Chow has extensive experience in networking and protocol design, network restoration, content switching, and network security.
c) How will the work specifically expand the university’s homeland defense and homeland security research capacity?
The project will support faculty members and graduate students directly supporting homeland security research. While we are addressing the security research issues from networking, distribute and web system perspectives, there is an increasing demand of integrating these different techniques and building a cohesive platform. The proposed work will also result in publications in the related fields, promoting the visibility of the university’s homeland defense and homeland security research activities. We also expect the preliminary results to play a key role in the follow-up external competitive proposals.
d) What will be the expected benefits to industry, government, and/or military?
By sharing secure information sharing technologies or providing easy to use systems with document tracking capability developed in the proposed project, it benefits industry, government and military organization are benefited to in the process of exchangeing sensitive information, plans, and procedures. Tools and techniques developed for secure ubiquitous access of sensor and web information may help speed up the introduction of sensor networks.
e)Who will work on the project?
One PI and one fundedone-funded GRAs (and 1-2 unfunded students) will work on the project.
Dr. Edward Chow is an Associate Professor in Computer Science at UCCS. He has been working on load balancing, content switching, resource allocation, network restoration, and network security issues. Recent research accomplishments include the development of proxy server based secure indirect routing protocol, secure DNS update with multiple path indirect routing entries, secure wireless authentication server supporting both PEAP and TTLS, secure groupware for first responders with instant messaging and group rekeying, and an autonomous anti-DDoS system. He has two US patents on distributed network restoration methods. He has published over 40 papers on networks and protocols.
f) What is the timeframe in which the proposed work will be accomplished and the level of effort for each of the participants?
We want to divide the execution of the proposal into three phrases and plan to conduct this project in three terms.
Timeframe / Task to be fulfilledPhase 1:
06/01/2004 – 08/31/2004 / Study and develop key and attributed certificate management for large scale information sharing and collaborative work.
Discuss with NISSC and USNorthcom to revise the tasks of the project.
Write and submit a DoD BAA proposal with the participation of NISSC staff.
Hold review/demo meeting at the end of Phase1.
Phase 2:
09/01/2004 – 12/31/2004 / Develop Infrastructure support for secure web-based collaborative applications.
Hold review and demo meeting at the end of Phase2.
Phase 3:
01/01/2005 – 05/31/2005 / Develop Ubiquitous computing for sharing sensor and web information.
Document and publish research results.
Based the research results, submit additional NSF or DoD proposal with NISSC as a partner.
Hold review meeting at the end of Phase3.
g) What related work currently exists in this field of research? Provide article references and/or Web sites where applicable.
See the reference section and the discussion in Sections 1 and 2.
h) What research funding do members of the research team currently possess and what grant proposal could be funded prior to or during this effort. If funding exists, state who has provided the funding (NSF, DARPA, etc.) and amount and duration of the funding.
The PI and has submitted 2 NSF proposals.
- C. Edward Chow, Xiaobo Zhou, and Terry Boult, “IMPACT: Improving Measurable Performance against Cyber Threats,” NSF Cyber-trust, 7/1/04-6/30/07, $866,294.
- Mark Wickert, C. Edward Chow, Rodger Ziemer, “Optimizing the Rate/Reliability Tradeoff Across Layers in Rapidly Deployable Networks,”NSF (CISE) ITR, 07/01/04 – 06/30/08, $1,004,930.
i) What tangible products (papers, grant proposals, technology transfer, etc) should be expected from this project?
A DoD BAA proposal will be submitted as a result of the phase I activities with NISSC participation. It is expected that the results of this project will be published in IEEE international conferences/workshops/journals.
It is expected that the results of this project will be used as preliminary results in NSF proposals in phase 3.
It is also expected that the resulted technology could be transferred to the industry.
j) Who might be future third-party sponsors of this work? List any specific research programs that are being targeted.
The PI intends to submit a proposal to Federal funding agencies, NSF or DoD, based on the preliminary research results developed from the proposed project. One targeted sponsor is NSF Division of Computer and Network Systems, Cyber Trust Programs (Spring 2005).
k) Does research involve human subjects, animal subjects, or drugs, controlled substances, radioactive materials, or recombinant DNA? If so, the appropriate campus committee approvals must be obtained.
NONE.
i) Has the researcher(s) been previously awarded a NISSC sponsored grant? If so, provide
The PI has received the following NISSC sponsored grants.
1. Admission control with adaptive resource management for defending degrading DDoS attacks
Xiaobo Zhou, C. Edward Chow
Aug 2003 – Dec 2003
Abstract
We developed an application-level service differentiation technique based on admission control and resource management for mitigating Degrading DDoS attacks.
Dates: interim report was submitted on Nov 3, 2003. Final report was submitted on Jan 31, 2004.
Published results:
Xiaobo Zhou, Yu Cai, Ganesh K. Godavari, C. Edward Chow, “An adaptive process
allocation strategy for proportional responsiveness differentiation on Web servers”, In
Proc. of 2nd International Conference on Web Services (ICWS), IEEE Computer Society,
San Diego, July 6-9, 2004.
2. Evaluating a cluster-based server platform in support of Intelligence/ Information Fusion and On Integrating System Support for Intelligence/Information FusionXiaobo Zhou, C. Edward Chow, Marijke AugusteijnFeb 2004 – May 2004
Abstract
The project aims to investigate resource management mechanisms in providing performance isolations and improvements for different applications in a cluster-based server platform, and hence evaluate the impact of the mechanisms in support of information/data fusion applications. The project also aims to investigate those challenging issues on how to exchange, verify, and correlate intelligence information for decision support, and how to allocate and coordinate sensors in different agencies for a set of tasks with different priorities.
Dates: interim report was submitted on May 3, 2004. The project is on-going. Final report is under preparation.
Published results: results are to be submitted to international conferences/workshops.
3. Secure Mobile Ad Hoc Network
Edward Chow
6/2/2003-8/31/2003
Abstract
A secure MANET system, called SMANET, was developed that accepts only those packets whose MAC addresses are in the Linux iptable firewall rules. Detailed iptable set up and the performance of the firewall are presented. SMANET is integrated with a simple intrusion alert system based on TCPDUMP utility.
Dates interim and final reports were submitted: 7/15/2003 (interim); 8/31/2003 (final).
Published results:
C. Edward Chow, Paul J. Fong, and Ganesh Godavari, “An Exercise in Constructing Secure Mobile Ad hoc Network (SMANET),” Proceeding of 18th ANIA 2004 Conference, Fukuoka, Japan, 3/2004.
4. SGFR: Enhancing the Security and Effectiveness of
Collaborative Software for First Responders
Edward Chow
6/2/2003-8/31/2003
Abstract
In this paper, we present the design and implementation of a secure groupware for first responders, called SGFR, that is capable of secure group chat, remote file download and remote display control. It integrated Jabber instant messaging system and Keystone group rekeying system. Users are authenticated through the use of digital certificates. Group key are issued when members are joined or leaves to ensure the security policy. The performance of SGFR is also presented. The system was first developed on Linux PC then ported to an IPaq PDA running Linux as a secure information delivery platform.