Wireless Networking and Next Generation Internet Architecture Research
at Washington University in Saint Louis

The Computer Science and Engineering (CSE) Department at Washington University in Saint Louis has a very aggressive program of research in Networking and Telecommunications. We are participating in many industry forums such as WiMAX Forum, IEEE 802, Internet Engineering Task Force (IETF), ANSI, International Telecommunications Union (ITU) and Telecommunications Industry of America (TIA). We collaborate with industry to ensure that we are working on relevant problems of current interest and that our solutions are adopted by the industry.

This page describes the research projects lead by Professor Raj Jain.

Current Research Projects:

  1. Internet 3.0: Architecture for the Next Generation Internet
  2. Next Generation Wireless Networks
  3. Communication and Modeling for Green Buildings
  4. Communications for Emergency Situations
  5. Resource Management in Wireless Networks
  6. Mobile Video Modeling
  7. Network Security
  8. Congestion Control and Traffic Management
  9. TCP Persistence
  10. Energy Efficient Protocols

1. Internet 3.0: Architecture for the Next Generation Internet

The key trend driving the growth of Internet over the last decade is the profusion of services over the Internet. Google, Facebook, YouTube and similar services form the bulk of the Internet traffic. Cloud computing and proliferation of mobile devices has lead to further growth in distributed services over the Internet. As part of our Internet 3.0 project, we are currently developing an open service delivery network (openSDN) architecture. openSDN will allow networking equipment companies to develop openSDN routers that will allow many application service providers (ASPs) to share a network and achieve the features required for widely distributed services, such as, load balancing, fault tolerance, replication, multihoming, and mobility that are customized for their application.

The proposed architecture is evolutionary in the sense that it can coexist and is backward compatible with the current Internet. It can be used both for services that reside on multiple virtual machines inside a cloud as well as those that are distributed over multiple clouds. The key component underlying openSDN is a rule-based delegation mechanism that allows application services to specify a set of rules for handling its traffic. These may include rules for selecting the instances among multiple replicas of the server, how to load-balance among different instances, what to do under network and instance failure (due to security attack or hardware malfunction), etc.

openSDN cleanly separates control and data planes. Control plane provides a secure interface for registration and distribution of specific rules, and translates the service rules to forwarding rules that are then sent to the data plane, which enforces these rules. All openSDN-aware entities have names, IDs and locators that help maintain multiple instances of services and allow the services to move transparently inside a cloud or among many clouds. An innovative naming system is used to resolve entity names to IDs. The naming system is designed to be backward compatible with the current domain name service (DNS) so that legacy hosts and services queries are resolved correctly.

Applications of openSDN include Private WANs, datacenter networks, cloud services, telecom services, online ervices, distributed application networks, science, educational and research computing, and defense networks.

This project is funded partly by NSF and was funded in the past by Intel Corporation.

Related Recent Publications:

Next Generation Wireless Networks

Many of the concepts of Internet 3.0 apply to wireless networks also.

Communication and Modeling for Green Buildings

Commercial and residential buildings are a major source of energy consumption. A number of products are appearing on the market that allow monitoring and control of various energy consuming devices. However, these devices from different manufacturers are often not able to communicate with each other because they are either not networked or use proprietary protocols. Also, the buildings often don't have network wiring at locations where energy consuming devices are usually located. A fully networked energy monitoring and control system is required to realize the potential energy savings while maintaining comfort for the occupants.

In collaboration with the Department of Computer Science and Department of Energy, Environmental and Chemical Engineering at Washington University, we plan to develop methods to monitor, model, and control energy usage in commercial and residential buildings. In particular, we want to develop easy methods to collect and analyze usage data. One of the problems in collecting the data is to network various monitoring devices. Although it is possible to network all monitoring equipment, this is feasible only in new buildings. Existing buildings may not have networking facilities or network wiring. Wireless communication is sometimes feasible in smaller distances. For larger distances we need more innovative networking techniques such as networking via electrical wires.

This is a new project funded by Washington University's I-CARES program.

Communications for Emergency Situations

Japan has a cellular Early Earthquake Warning (EEW) system in addition to the usual radio and television based systems. United States has only radio and television based emergency warning systems which are inadequate since most people do not listen to radio or watch television 24 hours a day (particularly those at work places). A cellular system is required since cell phone has now become the only communication device that can reach most of the population 24 hours/day 7-days a week.

There are three goals of this project: First, we will study the cellular EEW of Japan and its use in the March 11th earthquake in Japan. Second, we will study the problems that were encountered leading to so many deaths and missing persons. Third, we will use such patterns and inputs from the field to help develop a better cellular EEW for USA that uses the best of Japanese system and avoids its problems. In particular, we would like to explore new cellular communication modes such as tower-less phone-to-phone direct communication that is possible with new WiFi equipped smart phones.

This is a new project funded by NSF.

Resource Management in Wireless Networks

IEEE 802.16e based WiMAX networks promise the best available quality of experience for mobile data service users. Unlike wireless LANs, WiMAX networks incorporate several quality of service (QoS) mechanisms at the Media Access Control (MAC) level for guaranteed services for data, voice and video. The problem of assuring QoS is basically that of how to allocate available resources among users in order to meet the QoS criteria such as delay, delay jitter and throughput requirements. IEEE standard does not include a standard scheduling mechanism and leaves it for implementer differentiation. Scheduling is, therefore, of special interest to all WiMAX equipment makers and service providers. We have developed and analyzed several scheduling mechanisms for WiMAX.

Most of the resource management studies require simulation. It is important to have some common features among these models so that their results can be compared. Working with the Application Working Group at the WiMAX Forum, we have developed a standard simulation methodology that describes the key features to be simulated, the method of simulating these features and various parameter values to be used. This system level methodology has been used in several public and commercial WiMAX simulation models.

Related Recent Publications:

Mobile Video Modeling

Proper workload characterization is important for analyzing resource management schemes. Video streaming is continuously acquiring a larger and larger share of Internet's traffic resulting in a need to have a reliable video traffic model. We have analyzed several video streams for mobile streaming and have developed a a simple model, which we call Simplified Seasonal ARIMA Model (SAM). This model represents most of the video streams very well. Our library of video traces is available for other researchers to use.

Related Recent Publications:

Network Security

Distributed denial-of-service attacks (DDoS) pose an immense threat to the Internet. The most studied solution is to let routers probabilistically mark packets with partial path information during packet forwarding, which is referred as Probabilistic Packet Marking (PPM). We have shown that random marking is sufficient to impede the victim from tracing the attackers. A simple enhancement based on IP path length distribution makes it harder for the victim.

Key predistribution is a popular technique for key distribution in sensor networks. We have developed two key predistribution based scheme for heterogeneous networks i.e., networks which consist of nodes which are stationary as well as highly mobile.

With the growth and acceptance of the Internet, there has been increased interest in maintaining anonymity in the network. Using traffic analysis, it is possible to infer who is talking to whom over a public network. We have developed a novel approach to hide the senders and the receivers of messages. Our protocol poses no bandwidth overhead when there is at least some traffic while posing minimal bandwidth overhead when there is no traffic at all.

We have developed Air to Air Communication (AAC) - a wireless protocol designed for communication among airplanes as well as airplanes and control centers. AAC enables the broadcast of emergency and surveillance information such as realtime video over the network even in presence of adverse conditions such as coordinated terrorist attacks. AAC has the potential to significantly enhance the security of the homeland by closely monitoring the airplane which, if hijacked by terrorists or criminals, could be used as weapons.

Related Recent Publications:

Congestion Control and Traffic Management

Ethernet is replacing the traditional storage networking technologies like Fiber Channel and Infiniband in Datacenters. The key feature of these traditional technologies that make them suitable for datacenter is their low-loss low-delay operation. Consequently IEEE 802.1 standards committee is developing new specification for congestion management for Ethernet in datacenter networks.

We have developed an explicit rate control framework for Ethernet applications. The framework guarantees zero packet drops at the congested switch and fast convergence to fair and stable state.

Related Recent Publications:

TCP Persistence

Mobile applications often get disconnected because TCP times out when a user moves from one location and reconnects at another location. This happens even with the use of Mobile IP since Mobile IP hides the IP address change from TCP but does nothing to prevent it from timing out. In our proposed PErsistent TCP using Simple freeze (PETS) framework, we combine TCP freeze and Mobile IP to prevent TCP from disconnecting during mobile operations.

Related Recent Publications:

Energy Efficient Protocols

While broadcasting is a very energy-expensive protocol, it is also widely used as a building block for a variety of other network layer protocols, particularly in sensor networks. Therefore, reducing the energy consumption by optimizing broadcasting is a major improvement in heterogenous sensor networking. Our QoS Geometric Broadcast Protocol (QoS-GBP) is a distributed algorithm where nodes make local decisions on whether to transmit based on a geometric approach.

Related Recent Publications:

Collaboration:

In addition to traditional sources of research funding, our research in the past has also been sponsored by: Intel Corporation, Huawei, and Boeing Corporation.

Students

:
Chakchai So-In Abdel Karim Al-Tamimi Subharthi Paul Jianli Pan
Chakchai So-In Abdel-Karim Al Tamimi Subharthi Paul Jianli Pan
Back to Raj Jain's Home Page