Past and present projects in the NMSL
  • IMJ - Interactive Multimedia Jukebox
  • MControl - MBone Stream Controller
  • MHealth - Multicast Tree Visualization and Monitoring
  • Mwalk - Analyzing Properties of Multicast Trees
  • SDR Monitor - SDR Monitoring Effort
  • MANTRA - Monitoring and Analysis of Traffic in Multicast Routers
  • MRM - Multicast Reachability Monitor
  • pIP - Pseudo-IP
  • Digital Classroom - A Next-Generation Digital Classroom
  • PAIRwise - An Open-Source Plagiarism Detection System
  • AutoCap - An Open-Source Tool For Automatic and Reliable Captioning

Interactive Multimedia Jukebox
Straightforward, one-way delivery of audio/video through television sets has existed for many decades. In the 1980s, new services like Pay-Per-View and Video-on-Demand were touted as the ``killer applications'' for interactive TV. However, the hype has quickly died away leaving only hard technical problems and costly systems. As an alternative, we propose a new jukebox paradigm offering flexibility in how programs are requested and scheduled for playout. The jukebox scheduling paradigm offers flexibility ranging from complete viewer control (true video-on-demand), to complete service provider control (traditional broadcast TV).
MBone Stream Controller
The quantity of video and audio multimedia available over the Internet has been increasing at a dramatic rate over the past few years. One of the primary reasons is the close ties between the World Wide Web (WWW) and the scheduling/playout of audio and video programming. The convergence of audio/video and the WWW has created the potential for many compelling new services. One example is VCR-style functions for networked audio/video, which are traditional playout control functions that customers are familiar with including pause, rewind, fast forward, etc.
Multicast Tree Visualization and Monitoring
MHealth is a graphical, real-time multicast tree visualization and monitoring tool. MHealth attempts to draw a multicast tree topology for a given MBone group with the statistics available at a single end system. Two primary data sources are used to view and and monitor the tree structure: Real Time Control Protocol (RTCP) Packets and mtrace (a multicast traceroute tool). MHealth parses RTCP packets to build sender and receiver lists and uses mtrace to determine the route. Receivers are added into the tree as quickly as they can be traced. Loss is visually displayed in the form of both packet loss statistics by hop from mtrace, and color coding of the sender and receiver nodes by loss from both mtrace and the RTCP packets.
Analyzing Properties of Multicast Trees
Mwalk is a collection of java applications and perl scripts which re-create a global view of a multicast session from mtrace and RTCP logs. Once the session "tree" is constructed, a number of plug-in classes, visitors, can be employed to gather statistics while the tree is traversed or "walked". New visitors can easily be created by extending existing classes, and visitors can be combined to setup state within the tree that is later collected and analyzed.
SDR Monitoring Effort
Sdr Global Session Monitoring Effort is an effort to track, manage, and present information about the availability of world-wide sdr sessions. The basic idea is that if a number of people in topologically and geographically diverse places around the world send email with their sdr cache entries, a WWW page can be put together that says what sdr entries are being seen by what parts of the world. This type of service will provide all sorts of useful functionality. For example, if someone is trying to advertise a session they can check the "sdr-monitor" WWW page to see what percentage of sdr-monitor sites around the world see their session.
Monitoring and Analysis of Traffic in Multicast Routers
MANTRA is an attempt to develop a comprehensive tool which collects information at the router level and processes it online to give a current snapshot of the MBone. The tool can be used by router administrators as well as other MBone users to get a live status report of various local and global views of multicast groups. Thus, MANTRA is aimed at providing the capability of monitoring the status of various multicast groups and for providing an online analysis of their behavior.
Multicast Reachability Monitor
The Multicast Reachability Monitor (MRM) protocol has been designed to assist in the detection and isolation of network faults related to the delivery of multicast traffic. In particular, management functions offered by MRM are specifically designed to monitor routing operation, and assist in the investigation of routing anomalies and connectivity problems. The justification for MRM as a new protocol has followed the ``necessary and sufficient'' premise. MRM is being developed because it is necessary when comparing its functions to those offered by alternatives like the Real Time Control Protocol (RTCP) and the Simple Network Management Protocol (SNMP). Furthermore, MRM is being developed because it is sufficient in providing the functions needed by its target class of applications. Using this reasoning, MRM will offer functions and provide multicast traffic management that no other protocols currently offer.
In the near future people will be able to move freely and still have seamless network and Internet connectivity. One implication of this new style of interaction is an increased diversity in the types of devices which will be used to maintain connectivity. While many of the Internet protocols have proven successful and long-lived in traditional networks, we believe they will be inappropriate for use in communicating among limited-capability devices. Our proposed protocol, called Pseudo-IP, is designed to operate among devices in the farthest branches/leaves of an intranet while providing inter-network connectivity with other clouds and with the existing IP-based Internet infrastructure. Pseudo-IP will accommodate a wide range of devices with varying power, processing, and communication capabilities, while supporting a variety of applications.
Digital Classroom
The impact of technology on education is becoming more apparent everyday. Researchers have predicted that the use of technology in education will profoundly affect the learning process. The goal of our digital classroom is twofold. First, we establish the infrastructure to support the use of technology in learning. Second, we use our infrastructure as a testbed to develop news ways to integrate technology into the learning environment.
As the Internet becomes an indispensable part of university education, faculty members are finding new reasons for concern about plagiarism. Long a problem not easily addressed, the challenge of combating Internet plagiarism makes the task even more daunting.

PAIRwise offers instructors a tool that quickly and reliably deters and identifies likely instances of plagiarism. It is an automated, web-based system that compares students' work with one another and that searches the open Internet for matches with students' work. PAIRwise is an easy-to-use tool that identifies likely plagiarism while leaving substantive judgments about what is appropriate to instructors themselves.

Our AutoCap system automatically synchronizes pre- segmented transcripts with video. These transcript segments along with the time they were spoken, are combined to form captions. Unfortunately, no SRS in its normal configuration can achieve the accuracy necessary to synchronize transcripts without introducing an unreasonable amount of error for some or all of the captions. AutoCap overcomes this limitation by creating a better language model for use with the SRS that increases the accuracy to the necessary level. The role of the SRS is to collect as many groups of contiguous words, called utterances, as well as the time each word in the utterance was spoken. Once the SRS has collected all of these utterances, AutoCap creates captions by aligning the utterances to the transcript. For those words that are not recognized, AutoCap estimates when the words were spoken along with an error bound that gives the content creator an idea of caption accuracy. The result of this process is a collection of accurately time-stamped captions that can be displayed with the video.
updated 07.27.05