The multimedia and networking lab has a number of ongoing projects, that typically involve a mixture of analysis and experiments. The experiments are often carried out using a testbed consisting of multiple routers, hubs, and switches from different vendors. In particular, the testbed has been built using equipment generously donated by 3COM, IBM, and Lucent. Most of the projects currently carried out in the lab are supported through NSF grants, as well as through additional support from Sprint Labs and Nortel Networks. Brief outlines of several of the projects currently carried out in the lab follow.  
 
Evaluation of the performance and robustness of some of the new Internet service models such as those based on the Expedited Forwarding and the Assured Forwarding models.
  This work is based on both analysis (mostly simulations in this case), and experimentation on our testbed. In particular, all the routers in the testbed support various Quality-of-Service (QoS) capabilities that will be investigated in terms of support for the Differentiated Services model. In addition, a connection to the QBone (a QoS enabled portion of the new Internet2 network) will be used to complement the local testbed measurements with a wide area experiments. Of particular interest is the impact that network induced perturbations can have on conformance checks performed at boundaries between provider domains. This work was supported in part by a grant from Nortel Networks and by NSF grant ANI-9906855.

Related Publications

R. Guerin and V. Pla. "Aggregation and Conformance in Differentiated Service Networks: A Case Study." Computer Communication Review, Vol. 31, No. 1, January 2001, pp. 21-32. (Short version presented at ITC Specialist Seminar on IP Traffic Modeling, Measurement and Management, Monterey, CA, September 2000.)

 
 
Evaluation of the relation between application and network level performance.
  This work aims at a better understanding of how changes in network performance (and service parameters) affect the performance seen by applications and is supported by NSF grant ANI-9906855. One suite of experiments involves passing packet video streams through policers with different combination of parameters, and evaluate (quantitatively and qualitatively) the evolution of application level performance. The quantitative evaluation of video quality is done in collaboration with The Institute for Telecommunications Science using a tool they have developed. We plan on repeating and extending these experiments over the wide area QBone testbed. Testing over the QBone infrastructure is done in collaboration with members of the PennNet and Computing departments of the University of Pennsylvania (see also their Internet2 related activities) and with researchers at iCAIR and the IBM T.J. Watson Research Center.
Another set of experiments carried out with collaborators at AT&T Research, Florham Park, NJ, focused on the use of aggregate and "non-intrusive" performance measures for estimating the actual throughput experienced by TCP based applications. Non-intrusive refers to measures that are readily available from routine network monitoring, e.g., from routers MIBs, and do not require any flow specific awareness. The work involved the development of models (modification of existing models) that were capable of accurately predicting TCP throughput on the basis of such information, and the evaluation of their accuracy using both simulations and testbed experiments.

Related Publications

W. Ashmawi, R. Guerin, S. Wolf, and M. Pinson, "On the impact of policing and rate guarantees in Diff-Serv networks: A video streaming application perspective." In Proceedings of ACM SIGCOMM 2001, San diego, CA, August 2001. (A slightly extended version).

M. Goyal, R. Guerin, and R. Rajan, "Predicting TCP Throughput From Non-invasive Data."In Proceedings of INFOCOM'02, New York, NY, June 2002.

Shu Tao, Kuai Xu, Antonio Estepa, Teng Fei, Lixin Gao, Roch Guerin, Jim Kurose, Don Towsley, and Zhi-Li Zhang. "Improving VoIP Quality through Path Switching". Under submission.

Shu Tao and Roch Guerin, "Application-Specific Path Switching: A Case Study for Streaming Video". To appear in Proceedings of ACM Multimedia, New York, October 2004.

Shu Tao, Kuai Xu, Ying Xu, Teng Fei, Lixin Gao, Roch Guerin, Jim Kurose, Don Towsley, and Zhi-Li Zhang, "Exploring the Performance Benefits of End-to-End Path Switching". To appear in Proceedings of IEEE ICNP, Berlin, Germany, October 2004. A longer version is here.

Shu Tao, Kuai Xu, Ying Xu, Teng Fei, Lixin Gao, Roch Guerin, Jim Kurose, Don Towsley, and Zhi-Li Zhang. "Exploring the Performance Benefits of End-to-End Path Switching". In Proceedings of ACM Sigmetrics/Performance (Extended Abstract), New York, June 2004.

Shu Tao and Roch Guerin. "On-line Estimation of Internet Path Performance: An Application Perspective". In Proceedings of IEEE Infocom, Hong Kong, March 2004.

Xiaoxiang Lu, Shu Tao, Magda El Zarki, and Roch Guerin. "Quality-Based Adaptive Video Over the Internet". In Proceedings of CNDS, Orlando, FL, January 2003.

 
 
Reservations' Aggregation for Improved Scalability
  This work aims to study the problem of what is the best way to aggregate reservations in order to minimize the amount of state kept and processing that needs to be performed by backbone routers and therefore, to achieve the best aggregation trade-off. It is also focused in finding the optimal quantity of resources to request. The research involves the development and evaluation of efficient aggregation algorithms primarily by means of simulations in environments that make
use of Internet-like topologies.

Related Publications

R. Sofia, R. Guerin, and P. Veiga, "An Investigation of Inter-Domain Control Aggregation Procedures", a short version appeared in ICNP 2002, ESE Department, University of Pennsylvania.

R. Sofia, R. Guerin, and P. Veiga, "SICAP, a Shared-segment based Inter-domain Control Aggregation Protocol", Technical Report, ESE Department, University of Pennsylvania, October 2002. A short version appeared in HPSR'03, Italy, June 2003.
 
 
Aggregate versus Individual QoS.
  This work is supported through NSF grant ITR-0085930 (Scable Quality-of-Service Control for the Next Generation Internet)and is aimed at developing a better understanding of the relations that exist between QoS provided as some aggregate level, e.g., a service class as in the Differentiated Services model, and the actual QoS that individual users experience. Of particular interest are models that allow explicit evaluation of individual QoS measures, and their use in identifying characteristics of user traffic that can result in significant differences between individual and aggregate QoS measures. Initial work on this topic has led to the development of models that allow the evaluation of the loss probability experienced by individual connections and when and why it differs from the aggregate loss probability. The environment that is assumed consists of a single FIFO queue where all the individual users belonging to the same service class are multiplexed.

Related Publications

Y. Xu and R. Guerin, "Individual QoS versus aggregate QoS: A loss performance study."In Proceedings of INFOCOM'02, New York, NY, June 2002.

Ying Xu and Roch Guerin, "Individual QoS versus Aggregate QoS: A loss performance study" To appear in IEEE/ACM Transactions on Networking

Ying Xu and Roch Guerin, "On Evaluating Loss Performance Deviation: A Simple Tool and Its Practical Implications", Proceedings of 2nd International Workshop on QoS in Multi-service IP Networks (QoS-IP 2003), Milano, Italy, Feb. 2003
 
 
QoS extensions to IP routing protocols and traffic engineering applications.
  This is work has two components. The first builds on earlier projects aimed at understanding the cost-performance trade-off of QoS routing and an implementation of QoS routing capabilities in the context of the OSPF protocol. One extension currently being considered includes support for multiple areas in the context of OSPF and more generally issues associated with inter-network QoS routing. The second component of this work targets traffic engineering applications, i.e., the optimization of traffic distribution on a given network topology. The problems under investigation include how to best group traffic flows to minimize the number of distinct paths to establish while minimizing the impact on overall network performance, how this grouping affects both long term and short term performance, as well as evaluating traffic engineering solutions that can provide resilience to route changes caused by link or node failures. This work is supported in part by NSF grant ANI-9902943 and by a gift from Sprint ATL, and is carried out in collaboration with Christophe Diot and his group at Sprint ATL.

Related Publications

A. Sridharan, S. Bhattacharyya, C. Diot, R. Guerin, J. Jetcheva, and N. Taft. "On The Impact of Aggregation on The Performance of Traffic Aware Routing." In Proceedings of ITC'17, Brazil, December 2001

R. Guerin and A. Orda. "Computing Shortest Paths for Any Number of Hops." IEEE/ACM Transactions on Networking, Vol. 10, No. 5, pp. 613-620, October 2002 .

Ashwin Sridharan, R. Guerin, C. Diot, "Achieving Near Optimal Traffic Engineering Solutions in Current OSPF/ISIS Networks", Proceedings of INFOCOM 2003, San Fransisco, USA.

Ashwin Sridharan, Sue. B. Moon and C. Diot, "On the Correlation between Route Dynamics and Routing Loops", Proceedings of IMC 2003, Miami, Florida, USA.

Ashwin Sridharan, Roch Guerin, C. Diot, "Achieving Near Optimal Traffic Engineering Solutions in Current OSPF/ISIS Networks", accepted for publication in IEEE/ACM Transactions on Networking.
 
 
Routing and Scheduling Support for Advance Reservation.
  The routing aspect of this works targets the design and implementation of an infrastructure to support the computation of routes for advance reservations. It spans both the acquisition and maintenance of the necessary topology information, i.e., through interactions with existing routing protocols, and the design and implementation of efficient path selection algorithms together with their associated data structures. Or particular interest, is the design of an extended topology database that can efficiently store load information over time, and that can be easily accessed and modified by path selection algorithms. On the scheduling front, the work explores the development of a number schedulers aimed at increasing the flexibility with which bandwidth can be distributed across users with different types of reservations. One such direction involves supporting bandwidth guarantees with preemption capability. Another direction involves provision of delay guarantees while allowing for the transmission of excess traffic. Some of this work is supported in part by NSF grant ANI-9902943 and by a grant from Nortel Networks.

Related Publications

Y. Huang, R. Guerin and P. Gupta "Supporting Excess Real-time Traffic with Active Drop Queue", in Proceedings of TC'18, Berlin, August 2003.

Y. Huang and R. Guerin, "A Simple FIFO-Based Scheme for Differentiated Loss Guarantees", in Proceedings of IWQoS 2004, Montreal, Canada, 2004.

 
 
Resource Management in Wireless Networks.
 

Wireless networks are constrained in critical resources like bandwidth and power. The future success of large scale wireless applications will largely depend on how well these resources are managed. The objective of the projects in this area is to meet quality of service requirements through efficient exploitation of these limited resources. These projects involve a combination of analysis and experimentation. Parts of this research are conducted in collaboration with Prof. Sanjeev Khanna at Computer Science department, Prof. Vijay Kumar at the GRASP laboratory at University of Pennsylvania and Prof. Leandros Tassiulas at University of Maryland. We are currently supported by NSF grants ANI 6010793 and CISE 6059276, and by additional support from CECOM.

Two of the main areas we are looking at are:
Management of Bandwith: At present wireless links have bandwidth less than 10 Mb/s. However, next generation wireless applications envision simultaneous transmission of several multimedia streams. A typical video stream consumes 1-2 Mb/s. In addition wireless transmission suffers from channel errors and interference from other communications in vicinity. The research challenge is to appropriately time-share the transmission of different streams so as to guarantee the desired quality of service. Towards this end, we are investigating scheduling and routing strategies which attain desired bandwidth allocation objectives while satisfying the resource allocation constraints.
We have developed a collision resolution technique which infers the level of congestion from energy measurements and tunes the packet transmission procedure accordingly. This strategy outperforms the current collision resolution approaches in bandwidth utilization and power efficiency.
We are investigating scheduling strategies which attain fair allocation of bandwidth in wireless adhoc networks. The objective is to develop distributed scheduling policies which attain fair allocation while remaining oblivious to global topology and traffic parameters.
Research is underway in developing scheduling and routing strategies for other bandwidth allocation objectives in wireless networks.

Related Publication:

L. Tassiulas, S. Sarkar, "Maximin Fair Scheduling in Wireless Networks" In Proceedings of INFOCOM'02, New York, NY, June 2002.

S. Khanna, S. Sarkar, I. Shin, ``An Energy Management Based Collision Resolution Protocol,'' Technical Report, Electrical Engineering Department, University of Pennsylvania, 2001, Submitted for Publication.

S. Sarkar, K. N. Sivarajan, "Channel Assignment Algorithms Satisfying Cochannel and Adjacent Channel Reuse Constraints in Cellular Mobile Networks." Proceedings of INFOCOM 1998.

S. Sarkar, K. N. Sivarajan, "Fairness in Cellular Mobile Networks." Proceedings of 34th Annual Allerton
Conference on Communication, Control and Computing, August 1996.

S. Sarkar, K. N. Sivarajan, "Hypergraph Models for Cellular Mobile Communication Systems." IEEE
Transactions on Vehicular technology, May 1998.
Management of Power: Small handheld devices like PDA s will constitute the nodes of the next generation wireless adhoc networks. These nodes have limited battery power, and the scenario is not likely to change unless there is a significant breakthrough in battery technology. Thus power management is an important research issue for next generation wireless applications. The aim of this project is to develop battery management and scheduling and routing strategies which increase the lifetime of a network. Our initial focus has been to design an optimal framework for scheduling of wireless transmissions which exploit the residual power characteristics of a battery so as to maximize the lifetime of a single wireless node. Ongoing research is directed towards incorporating routing and scheduling decisions in the above framework for application in a multi-node network scenario.

Related Publication:

M. Adamou, S. Sarkar, "A Framework for Optimal Battery Management for Wireless Nodes." In Proceedings of INFOCOM 1999, Accepted for Publication in IEEE Transactions on Information Theory.