AFOSR MURI Project Progress Report (July 2002)

PI Information:

Tatsuya Suda
Department of Information and Computer Science
University of California, Irvine
Irvine, CA 92697-1875
Phone: 949-824-7403
FAX: 949-824-2886
E-mail: suda@ics.uci.edu
Web: http://netresearch.ics.uci.edu/

Summary of the Proposed Work

The projects that the PI originally proposed in the MURI proposal are to investigate scalable, real-time video multicast protocols and also a new network architecture called the Bio-Networking Architecture and its associated protocols. The primary goal of the first project is to provide efficient and scalable multicast of high quality, real-time video to a large number of destinations. The primary goal of the second project is to provide scalable, adaptive and survivable network applications and protocols by adopting biological concepts and mechanisms.

As the MURI projects progress, the PI expanded his MURI efforts to include other key research areas in protocol designs, including investigation of protocols for distributed peer-to-peer systems and protocols for wireless and small-device networks. The goal of the investigation of peer-to-peer system protocols is to develop distributed protocols to satisfy queries for particular network objects, taking into consideration the distributed, dynamic and large scale characteristics of a peer-to-peer system. The goal of the wireless small-device network protocols is to support mobility, ad hoc networks and disconnected operation with lower overheads and greater predictability. These new efforts are consistent with the vision and goals of the original MURI proposal.

With the support from AFOSR, the PI has made significant progress and research contributions. Some evidence of the impact of the PI’s MURI research are listed below.

• BBC News featured an article that overviews the motivation and vision of the Bio-Networking Architecture [BBC00]. In this article, adaptive object behaviors and network applications in the Bio-Networking Architecture are described.
• Key protocol designs in the Bio-Networking Architecture are currently being discussed as a possible reference architecture at the Super Distributed Objects group of the Object Management Group (OMG), the largest standard making bodies for object oriented software technologies [SAS01, SSS02].
• NS Solutions Corporation, one of the largest system integrators in Japan, is interested in marketing the scalable, real-time video multicast protocol that the PI has developed [AVS01,VAS00].
• One of the papers that the PI published [INMSA02b] on the Bio-Networking Architecture and its associated protocols received the best paper award at the 2002 IEEE SAINT (Symposium on Applications and the Internet) conference. This is a major IEEE conference specializing in Internet applications with ten technical paper sessions, three panel sessions and 4 workshops.

Progress in understanding and developing protocols has been evidenced through achievements in various projects that the PI investigates as a part of his MURI efforts, which include research within the Bio-Networking Architecture, peer-to-peer systems, and wireless and small device networks. In the Bio-Networking Architecture, the PI has obtained deep understanding of how biological concepts and mechanisms can help network application technology and protocols, and shown that this new paradigm of applying biological concepts and mechanisms is feasible and promising. In peer-to-peer systems, the PI has addressed how discovery protocols may be adjusted to perform in a distributed and large-scale manner. In wireless and small device networks, aspects of dynamic and unpredictable environments are applied in protocol designs. Through work in these areas, the PI has advanced research in the breadth of areas necessary for the design of networking protocols.

To date, the PI has produced a number of artifacts, marking advances in each of these areas.

• The PI has produced designs of new protocols: scalable, real-time video multicast protocols; protocols necessary to implement the Bio-Networking Architecture; protocols for discovery in peer-to-peer systems; and various new protocols for wireless and small-devices networks.
• The PI has produced simulation models and simulators, in order to demonstrate characteristics of the simulated protocols and to aid in the design and evaluation of the various proposed protocols.
• A software implementation of the Bio-Networking Architecture and its associated protocols is in progress in order to empirically evaluate the Bio-Networking Architecture protocols.
• A large number of publications describing various achievements in the PI’s MURI projects. These include 3 standards documents, 2 book chapters, 5 journal papers, 9 workshop papers/conference abstracts, 12 refereed conference papers, 4 non-refereed conference papers, 2 technical reports, and 6 papers in submission.

Please note that, due to the large number of publications within each project, the number of publications listed in the following descriptions is limited to a few.

This section briefly summarizes research activities and results obtained to date since the beginning of the MURI projects, May of 2000.

Scalable, Real-time Video Multicast Protocols

In this research, the PI has proposed and investigated a novel protocol, called Source-Adaptive Multi-layered Multicast (SAMM), to control congestion caused by the real-time video multicast over the Internet [AVS01, VAS00]. SAMM relies on the exchange of feedback between end systems (a video source and multicast destinations) and rate adjustment at a video source. In SAMM, multi-layered video encoding is deployed, and raw video data is encoded into one or more layers of differing priority. Video sources adapt the number of video layers they generate as well as the transmission rate of each video layer in response to congestion feedback from the network and receivers. Using simulations that incorporate multi-layered video codecs, the PI has demonstrated that SAMM exhibits better scalability and responsiveness to congestion than existing protocols. Accomplishments in this work include design of the SAMM protocols, a simulator to evaluate the proposed protocols and simulation results confirming the performance of the proposed protocols.

The Bio-Networking Architecture

In the Bio-Networking Architecture, complex network applications emerge through interactions among multiple objects (i.e., service components). Since objects may be developed independently by different designers, it is important to ensure that independently developed objects can communicate and interact to collectively provide an application. The PI is currently developing a set of specifications (called Flexible Interface Definition) that defines the interfaces and protocols of each object in a flexible manner so that objects implementing different interfaces and protocols can communicate and interact. The PI has shown that Flexible Interface Definition increases the number and variety of network applications that objects can create. Accomplishments in this work include design and preliminary implementation of Flexible Interface Definition [FS02].

The PI is also developing the Bio-Networking platform, a middleware that provides reusable software components for deploying and executing objects (i.e., service components) and communication protocols between objects. Platform components abstract low-level operating and networking details (e.g. message marshalling/unmarshalling, I/O, concurrency, and network connection management) and provide a series of runtime services that objects frequently use for performing their services and invoking their protocols. Accomplishments in this work include the design and preliminary implementation of the Bio-Networking platform, as well as preliminary measurements showing the efficiency and scalability of the Bio-Networking platform [SS02a]. The design of the Bio-Networking platform and its protocols has been proposed to OMG for possible standards for Super Distributed Objects specifications [SFS02].

Peer-to-Peer Discovery Protocols

Peer-to-peer systems contain distributed objects (users, data, and services) with highly dynamic properties. For example, objects may dynamically enter or leave the peer-to-peer network, or attributes of the object may change. Discovery protocols in peer-to-peer systems define how to locate specific objects based on the content or attributes of the objects. The PI has developed and evaluated two distinct discovery protocols that perform discovery to locate objects with particular keywords. Both discovery protocols perform forwarding of queries along logical links (relationships) between the objects in a distributed manner.

In this first discovery protocol that the PI developed, discovery queries are forwarded based on keyword similarity and discovery history of relationships between objects [MS02]. In the second discovery protocol, a user’s (i.e., discovery originator’s) evaluation of the received discovery hits is used to guide forwarding of queries [S2M]. In both discovery protocols, relationships are weighted or selected in order to enhance the forwarding properties of the protocol. Accomplishments for this work include: designs for each discovery protocol; simulations demonstrating protocol efficiency, the quality of results returned, the robustness to dynamic environments and the scalability of these protocols.

Wireless Network Protocols

The PI has investigated key protocol design issues in both infrastructure based wireless networks and ad hoc wireless networks.

• Bandwidth Efficient Multicast Protocol in Ad Hoc networks. In this research, the PI has proposed and investigated a bandwidth-efficient multicast routing protocol for ad-hoc networks [OKS01]. The proposed protocol achieves low communication overhead (i.e., it requires a small number of control packet transmissions for route setup and maintenance). The proposed protocol also achieves high multicast efficiency (i.e., it delivers multicast packets to receivers with a small number of transmissions). Simulation results confirmed key features of the proposed protocol. Accomplishments in this work include design of the proposed protocol, and a simulator to evaluate the proposed and other ad hoc network protocols.

• Multipath Transmission Protocol for Smooth Wireless Handoff. The current mobile wireless protocols often fail to provide smooth handoffs due to packet rerouting and different amounts of available bandwidth in different wireless cells. In order to provide smooth handoffs for real-time video applications, the PI has proposed an end-to-end soft handoff protocol with multipath transmission. In the proposed protocol, the sender encodes video using a layered coding technique. Multiple paths are established from the sender to the mobile node (destination) when the destination is in a handoff area (i.e., an area that two wireless cells overlap). The sender transmits (duplicated) base layer video packets on all the paths to the destination to minimize the base layer video packet loss. Enhanced layer video packets are transmitted only through paths with additional bandwidth to increase the video throughput. The proposed protocol reduces the negative impact of the base layer video packet loss and increases the video throughput during handoffs. Accomplishments in this work include design of the proposed protocol, a simulation model for the proposed protocol, and the simulator that evaluates the proposed protocol and existing handoff protocols.

Protocols for Small-Device Networks in Homeland Security Scenarios

In order to provide efficient and uninterrupted communication services among victims, rescue team members, and law enforcement officers in disaster scenarios, it is important to restore connectivity among sub-networks that become segmented due to cable and communication equipment failures caused by disasters. It is also important to efficiently route information over these segmented sub-networks while taking into consideration the severely limited capabilities of the available communication devices (such as limited processing and transmission power capabilities of mobile phones that victims carry, or of sensors imbedded in building infrastructures).

In order to regain connectivity between segmented sub-networks, we consider two approaches: one approach involves deploying intelligent robot-type communication devices that autonomously move and place themselves in strategic locations, and the other approach involves deploying a massive number of simple communication devices (such as sensor type devices) scattered from, for instance, an aircraft flying over the disaster struck area. The PI is currently investigating protocols for communication between intelligent robot-type devices, as well as protocols for simple devices to relay information between segmented sub-networks. In considering such protocols, special attention is paid to the fact that simple devices may not be able to identify their locations in a precise manner and that some devices may be more critical than others in maintaining network connectivity. Accomplishments in this work include initial designs of protocols for simple devices to exchange coordinate information, and routing protocols for simple devices to relay information while considering the limitations of the devices.

This section summarizes the planned research activities and schedule of the on-going MURI research. In some of the MURI projects (scalable real-time video multicast protocols, peer-to-peer discovery protocols, and bandwidth efficient multicast protocol in ad-hoc networks), the PI has completed protocol designs and evaluation of the proposed protocols through simulations. For such projects, the PI plans to follow the research schedule listed below.

December 2002: Complete initial prototype implementations of the proposed protocols
July 2003: Complete refinement of the prototype implementations
July 2004: Complete empirical evaluation of the proposed protocols

Some of the MURI projects (The Bio-Networking Architecture, multipath transmission protocol for smooth wireless handoff, and protocols for small-device networks in homeland security scenarios) are at their beginning stages, and in such projects, the PI plans to follow the research schedule listed below.

Analytical Approach:

• December 2002: Complete building of mathematical models for protocols
• July 2003: Complete mathematical analysis of protocols
• December 2003: Complete evaluation of protocols through analytical research

Simulation Approach:

• April 2003: Complete investigation of various protocol components and options
• December 2003: Complete building of simulators to evaluate proposed protocols
• July 2004: Complete evaluation of protocols through simulations
• December 2004: Complete analysis of simulation results

Empirical Approach:

• July 2003: Complete high-level implementation designs of protocol components
• December 2003: Complete detailed implementation designs of protocol components
• July 2004: Complete prototype implementation of the proposed protocols
• December 2004: Complete empirical evaluation of the proposed protocols in relatively simple testbed environments and refining them based on the evaluation
• April 2005: Complete evaluation of the proposed protocols in more complex testbed environments

Scalable, Real-time Video Multicast Protocols:

[AVS01] C. Albuquerque, B. Vickers and T. Suda, "A Source Adaptive Multi-Layered Multicast Algorithm for Internet Video Distribution," the Springer-Verlag book entitled "Multimedia Internet Broadcasting", 2001.

[VAS00] B. J. Vickers, C. Albuquerque and T. Suda, "Source Adaptive Multi-Layered Multicast," the IEEE/ACM Trans. on Networking,Vol.8, number 6, pp. 720-733, December 2000.

[AVS00b] C. Albuquerque, B. Vickers and T. Suda, "Credit-Based Source-Adaptive Multi-Layered Multicast," Performance Evaluation (International Journal), No. 1-3, Vol.40, pp. 135--159, 2000.

The Bio-Networking Architecture:

[FMT02] K. Fujii, M. Imada, M. Matsuo and T. Suda, "Service Description in Ja-Net", submitted to the National Meeting, Information Processing Society of Japan, in Japanese, 2002.

[SS02a] J. Suzuki and T. Suda, "The Bio Networking Architecture: Its Platform Design and Platform Implementation," submitted to the International Symposium on Distributed Objects and Applications, 2002.

[OMS02] T. Otsuka, M. Matsuo, T. Suda, "Evolutionary Service Composition based on Patched Functional Components," submitted to the IEICEJ and ISPJ Forum on Information Technology, 2002.

[YSF] N. Yamashita, T. Suda and M. Fukushima, "An Equilibrium Model of a Self-organizing Network Architecture," in preparation for submission to a conference.

[SMINIOT02] T. Suda, M. Matsuo, T. Itao, T. Nakamura, M. Imada, T. Otsuka and S. Tanaka, "Ja-Net Adaptive Networking Architecture for Service Emergence," the Trans. of the Information Processing Society of Japan (IPSJ), Invited Paper, pp.616-622, in Japanese, June 2002.

[SINM01] T. Suda, T. Itao, T. Nakamura and M. Matsuo, "Adaptive Networking Architecture for Service Emergence," the Trans. Inst. Electronics Commun. Engineers of Japan (IECEJ), Invited Paper, Vol.J84-B, No.3, pp.310-320, in Japanese, 2001.

[INMSA02a] T. Itao, T. Nakamura, M. Matsuo, T. Suda, and T. Aoyama, "Adaptive Creation of Network Applications in the Jack-In-The-Net Architecture," Proceedings of the IFIP Networking 2002.

[INMSA02b] T. Itao, T. Nakamura, M. Matsuo, T. Suda, and T. Aoyama, "Service Emergence based on Cooperative Interaction of Self-Organizing Entities," Proceedings of the IEEE SAINT 2002, 2002.

[IKMS01] M. Imada, Y. Katayama, M. Matsuo and T. Suda, "A Method of Service Composition using Service Attributes," the IEICEJ Assurance Symposium, in Japanese, June 2001.

[ISA01] T. Itao, T. Suda and T. Aoyama, "Jack-in-the-Net: Adaptive Networking Architecture for Service Emergence," Proceedings of the Asian-Pacific Conference on Communications, 2001.

[ILS01] T. Iizuka, A. Lau and T. Suda, "A Design of Local Resource Access Control for Mobile Agent in PDA," Proceedings of the Asian-Pacific Conference on Communications, 2001.

[INIMSA01] T. Itao, T. Nakamura, M. Imada, M. Matsuo, T. Suda, and T. Aoyama, "The Model and Design of Cooperative Interaction for Service Composition," Proceedings of the IPSJ (Information Processing Society of Japan) DICOMO 2001 workshop.

[SS01] S. Song and T. Suda, "Security in Energy Level in Bio-Networking Architecture," the 3rd International Conference on Advanced Communications Technology, 2001.

[WS01] M. Wang and T. Suda, "The Bio-Networking Architecture: A Biologically Inspired Approach to the Design of Scalable, Adaptive, and Survivable/Available Network Applications," Proceedings of the IEEE SAINT Symposium, 2001.

[SS02b] J. Suzuki and T. Suda, "Adaptive Behavior Selection of Autonomous Objects in the Bio-Networking Architecture," Proceedings of the First Annual Symposium on Autonomous Intelligent Networks and Systems, May 2002.

[FS02] K. Fujii and T. Suda, "Loose Interface Definition: An Extended Interface Definition for Dynamic Service Composition," Proceedings of the First Annual Symposium on Autonomous Intelligent Networks and Systems, May 2002.

[NMIS02] T. Nakamura, M. Matsuo, T. Itao and T. Suda, "System Control in Ja-Net, a New Network Architecture with Service Emergence," IEICEJ Next Generation Networks workshop, December, 2001

[SS01a] S. Sugizaki and T. Suda, "Evolution Simulations for the Bio-Networking Architecture," IEICEJ Next Generation Networks workshop, December, 2001

[S01] T. Suda, "The Bio-Networking Architecture: A Scalable and Self-Organizing Network Architecture based on Biological Concepts," Santa Fe Institute Workshop on The Internet as a Large-Scale Complex System, Santa Fe, March 30, 2001.

[SS01b] J. Suzuki and T. Suda, "Bionet Project Overview: Applying Biological Concepts and Mechanisms for Designing Adaptable, Scalable and Survivable Communication Software," JSPS Workshop on Applied Information Technology for Science, 2001.

[SS01c] S. Song and T. Suda, "Security Considerations in the Bio-Networking Architecture," JSPS Workshop on Applied Information Technology for Science, 2001.

[ISNM01] T. Itao, T. Suda, T. Nakamura and M. Matsuo, "Adaptive Networking Architecture for Service Emergence," Proceedings of the Workshop on Highly Distributed Systems in the IEEE SAINT Symposium, 2001.

[IMS02] M. Imada, M. Matsuo and T. Suda, "A Service Composition Method Consisted of Cyber Entity," IEICE General Conference March, 2002.

[S01] T. Suda, "Jack-in-the-Net (Ja-Net) Architecture for Service Emergence," Proceedings of the Annual National Meeting of IECEJ, Sept. 2001.

[IMS01] M. Imada, M. Matsuo and T. Suda, "Service Composition based on Service Attributes in Ja-Net," IEICE IN Technical Group meeting, in Japanese, July 2001.

[ISNIM01] T. Itao, T. Suda, T. Nakamura, M. Imada and M. Matsuo, "Ja-Net: Adaptive Networking Architecture for Service Emergence," Proceedings of the Monthly Meeting of IECEJ, July 2001.

[SS01d] S. Song and T. Suda, "Security on Energy Level in the Bio-Networking Architecture," Technical Report, 01-11, Department of Information and Computer Science, University of California, Irvine, 2001.

[WS00] M. Wang and T. Suda, "The Bio-Networking Architecture: A Biologically Inspired Approach to the Design of Scalable, Adaptive, and Survivable/Available Network Applications," Technical Report, 00-03, Department of Information and Computer Science, University of California, Irvine, 2000.

Peer-to-Peer Discovery Protocols

[SIM] T. Suda, T. Itao, M. Masato, "The Bio-Networking Architecture; A Biologically Inspired Approach to the Design of Scalable, Adaptive, and Survivable/Available Network Applications," to appear in a book, The Internet as a Large-Scale Complex System, Princeton University Press.

[MS02] M. Moore and T. Suda, "Distributed Discovery in Peer-to-Peer Network," Proceedings of the First Annual Symposium on Autonomous Intelligent Networks and Systems, May 2002.

Wireless Network Protocols:

[OKS01] T. Ozaki, J. B. Kim and T. Suda, "Bandwidth-Efficient Multicast Routing Protocol for Ad-Hoc Networks," Proceedings of the IEEE Infocom, 2001.

[SS02c] J. Suzuki and T. Suda, "Middleware Support for Disaster Response Infrastructure," the IEEE First Workshop on Disaster Recovery Networks, June 2002.

Standard documents:

[SAS01] S. Sameshima, S. Arbanowski and J. Suzuki (ed.), "OMG Super Distributed Objects White Paper," Super Distributed Objects Domain SIG, Object Management Group, July 2001.

[SSS02] S. Sameshima, S. Steglich and J. Suzuki (ed.), "PIM and PSM for Super Distributed Objects," Request for Proposal, Super Distributed Objects Domain SIG, Object Management Group, OMG document number: sdo/02-01-04, January 2002.

[SFS02] J. Suzuki, K. Fujii and T. Suda, "The UCI Initial Proposal to SDO PIM," Super Distributed Objects Domain SIG, Object Management Group, June 2002.

Other Protocol Papers:

[ILS] F. Ishizaki, G. Lin and T. Suda, "On-line Sensitivity Queuing Analysis with Utilization of the Proportional Relationship," submitted to the IEEE Transactions on Automatic Control.

[AVS00] C. Albuquerque, B. Vickers and T. Suda, "Network Border Patrol: Preventing Congestion Collapse and Promoting Fairness in the Internet", submitted to the IEEE/ACM Transactions on Networking.

[HAOS01] D. Hong, C. Albuquerque, C. Oliveira and T. Suda, "Evaluating the Impact of Emerging Streaming Media Applications on TCP/IP Performance," The IEEE Communications Magazine, April 2001.

[AVS00] C. Albuquerque, B. Vickers and T. Suda, "Network Boarder Patrol," Proceedings of the IEEE Infocom, 2000.

[ALOS01] S. Ahn, M. Lee and T. Suda, "Topology Reconfiguration of an IP Network Embedded Over an ATM Network," Proceedings of the Asian-Pacific Conference on Communications, 2001.