Process Query Systems -- logo Dartmouth College Thayer School of Engineering logo
Process Query Systems -- left navigation image Process Query Systems -- left navigation background image
Process Query Systems -- menu divider image
  Main
     - Overview
     - Autonomic Computing
     - Network Security
     - Airborne Plume Tracking
     - Social Networks
     - Visual Tracking
     - Glossary Process Query Systems -- menu divider image
  Projects
     - PQSNet
     - Fish Tracking
     - Thermal Tracking
     - Plume Detection
     - Network Monitoring
     - ACTCOMM (external) Process Query Systems -- menu divider image
  Downloads
Process Query Systems -- menu divider image
  Publications
Process Query Systems -- menu divider image
  People
     - Current
     - Alumni Process Query Systems -- menu divider image
  

Process Query Systems -- Glossary

Glossary

  • Experiments:
    • An experiment is one multiple hypothesis tracker using one or more models to evaluate the likelihood (score) of possible tracks.

  • Hypothesis:
    • A conglomeration of tracks determined by PQSNet to support a prediction of future behavior.

  • Models:
    • A model (or process query) is a function that takes a track as the argument and produces a score as output. The higher the score, the more likely the modeled scenario is occurring.

  • Observations:
    • Individual pieces of data from one or more sensors. Ideally the output of a PQS is formatted such that it can be used as input observations for a second-level PQS. This essentially means that the conclusions published by a PQS can be used as observations for another PQS, conceptually allowing us to chain multiple Process Query Systems.

  • Process Query:
    • The act of applying a model to sensor data.

  • Score:
    • Measure of the likelihood that a predicted scenario is occurring. The scoring scheme depends on the model.

  • Sensors:
    • Sources of observations, in arbitrary form; for example, Snort or syslog.

  • Tracks:
    • A conglomeration of observations determined by PQSNet to represent a trend.

  • TRAFEN:
    • TRAcking and Fusion ENgine.

  • Snort:
    • An open-source, industry-standard intrusion detection system. Essentially a network sniffer with a large library of known attack "signatures" which can detect and act on those signatures in real time.

This research program is a part of the Institute for Security Technology Studies, supported under Award number 2000-DT-CX-K001 from the U.S. Department of Homeland Security, Science and Technology Directorate. Points of view in this web site are those of the authors and do not necessarily represent the official position of the U.S. Department of Homeland Security or the Science and Technology Directorate. Process Query Systems, LLC.

© Copyright 2005-2007 Trustees of Dartmouth College.
All rights reserved.