Time-Critical Detection of Anomalous Behaviour
Ard Huizing · TNO Repository · 2008
Terrorist attacks such a suicide bombings and detonation of improved explosive devices are a worldwide problem that caused many casualties and economic damage. To prevent these attacks, human intelligence gathering, social network analysis and data mining of databases containing financial transactions and telephone communication records are employed. This pre-emptive approach is, however, not fail-save and cannot predict and preclude all terrorist attacks. Therefore, procedures and techniques must be used that can identify imminent terrorist attacks by detecting anomalous behaviour of people, vehicles or other objects before the attacks occur. The time-critical detection of anomalous behaviour that is described in this paper is based on the fusion of data from multiple sensors and a consistency check of the sensor reports that are associated with a subject or an object.