Downloads Logo

Downloads

On this page are downloads of data-sets, traces, source code, utilities and multimedia files that have been used in our research. If you download and use any of the below in any research, please let us (claypool@cs.wpi.edu) know. Most can be extracted via "gzip -dc filename.tgz | tar xf -" or via "unzip filename.zip". Have fun!

Index

Age of Mythology

We did some experiments measuring the effects of latency on user performance in Age of Mythology. You can download the maps we used as well as some network traces.

Used in:

Battery Power

We wrote some software that can measure the battery drain rate on a laptop PC. You can download the Visual C++ source code (with executable already made) here:

Command and Conquer: Generals

We did some experiments measuring the effects of latency on user performance in Command and Conquer: Generals. You can download some network traces here.

Used in:

Dynamic Class Based Thresholds (D-CBT)

D-CBT is an router queue management mechanism that extends RED by having three classes of traffic: TCP, UDP and tagged UDP (multimedia). It provides fairness among the classes of flows using dynamic thresholds for each class.

NS version:

Used in:

Linux version:

Used in:

Goddard Streaming Media System

We have design and implemente a streaming system (client and server) called Goddard for use in the NS simulator. Goddard is designed based on the behaviors of Real Networks streaming media and Windows Stream media, as observed in measurements studies from work at the Video Tracer project page.

Used in:

(Note, the system is named after Robert Goddard, the "Father of Modern Rocketry", and a WPI alumnus.)

Implicit Interest Indicators

Some of the research in our REFER group is looking at the correlation between user Web browsing actions (implicit interest) with explicit interest. We built a Web browser (called Curious Browser) that records scrolling, access time, mouse movement and mouse clicks, conducted user tests with a bunch of volunteers, and analyzed the gathered data. The Curious Browser and the dataset collected is available below.

Used in:

Madden Online NFL Football

We studied the effects of latency on Madden Football. We also captured some network traces, which you can look at here.

Used in:

MM-Flow

MM-Flow is a set of rate-based applications for NS that use AIMD congestion control built on top of UDP. They are designed for better-performing yet responsive multimedia applications. We've implemented them in NS. Note, these are included with the ns-2 tree if you download D-CBT above.

Used in:

MPEG Movies

Each movie clip was recorded from a VCR using a Video ATI All-in-one capture card into AVI format then converted into MPEG-1. Each clip is about 60 seconds long and 10 MBytes in size.

Used in:

MPFS Benchmark

The benchmark source code here can be used to evaluate the performance of MPFS file systems.

Used in:

NISTNet DropTail

By default the NISTNet emulator manages the router queue by dropping packets using the DRD algorithm. We implemented a simple drop tail implementation, modifying the kernel/knistnet.c source code file. You can download the source code below and read about compilation and installation on the NIST Net page, as they remain the same.

QFind

The results data we obtained (minus the ping distributions) from our QFind measurements are available here.

Used in:

Starcraft and Counter-strike

Network games are becoming increasingly popular, but their traffic patterns have received little attention from the academic research community. We have examined network traffic from Counter-Strike and Starcraft thus far. The Commview traces and a parsing tool are available here.

Used in:

Receiver Based Auto Rate (RBAR)

Receiver Based Auto Rate (RBAR), proposed by Holland et al. in "A Rate-adaptive MAC Protocol for Multi-hop Wireless Networks" (Proceedings of Mobile Computing and Networking, 2001), uses analysis of RTS frames to measure channel quality. RBAR receivers determine the highest feasible frame transmission rate that channel conditions can tolerate and notify the sender of the chosen rate via the CTS frame. Since RTS/CTS messages are sent to the AP, all wireless nodes become aware of the new transmission rate and set their backoff timers accordingly.

Unfortunately, this algorithm is not integrated into ns-2 releases. However, an RBAR simulation module for ns-2 2.1b7 is downloadable from the Rice Networks Group. We re-implemented RBAR in NS 2.27 and extended the physical layer parameters using the specifications of the Lucent OriNOCO wireless PC card. Our implementation can be found here with documentation.

By default, ns-2 provides a two-way ground propagation module. To more accurately simulate physical condition effects on RCAs, an additional ns-2 extension module was implemented that models Ricean (or Rayleigh) fading (Punnoose et al. in Proceedings Vehicular Technology Conference, 2000).

Selective Flooding Simulator

Selective Flooding is a QoS routing algorithm checks the state of the links on a set of pre-computed routes from the source to the destination in parallel and based on this information computes the best route and then reserves resources. We implemented Selective Flooding by extending a QoS routing simulator. That simulator is available here:

Used in:

SRP - A Selective Retransmission Protocol

Selective Retransmission Protocol (SRP) balances the potentially high loss found in UDP with the potentially high latency found in TCP. SRP uses an application-specific decision algorithm (based on an audioconference in this implementation) to determine whether or not to ask for a retransmission for a lost packet, adjusting the loss and latency to the optimum level for the application.

Linux version:

Used in:

Stats

Stats is designed to be a quick, simple to use program as a Unix filter to generate the summary statistics.

UDP Tools

UDP Ping and UDP Heartbeat are customized ping tools using application-layer UDP packets to provide delay and loss information. UDP Ping reports round-trip time information while UDP Heartbeat reports one-way link information. Both require applications to be run on the end-hosts.

Download source, README and executables (windows):

See also:

Unreal Tournament 2003

In studying FPS games under network degrdataions, we've looked at UT2003. We ran multiple full-length games with one player matched against two bots on a small standard map (DM-GAEL) with four different conditions of packet loss and latency. For each game, we captured all network packets for 120 seconds during the middle of the 5 minute match. You can get the tcpdump traces here:

Used in:

Video Crawler

We crawled the Internet from many carefully selected starting points in an effort to characterize vidoes stored on the Web. You can download the URLs we crawled and the video URLs we found.

Used in:

Warcraft III

We did some experiments measuring the effects of latency on user performance in Warcraft III. You can download the maps we used as well as some network traces.

Used in:

WBEST

WBest is a wireless bandwidth estimation tool designed for applications that requires fast convergence time and low intrusiveness, such as multimedia streaming applications. WBest employs packet dispersion techniques to provide capacity and available bandwidth information for the underlying wireless networks.

Used in:

WRAPI+

WRAPI+ is a tool to monitor wireless statistics, including received signal strength, transmitted frame count, and failed frame transmissions and acknowledgments on a Windows XP end hosts' IEEE 802.11b/g network device. WRAPI+ was built upon the freely available WRAPI C++ library.

Download (for Windows XP):

Used in:

See also: