Title :
Adaptive real-time training strategies for a trainable multi-sensor perimeter intrusion detection system
Author :
Bartek, Richard J. ; Sanders, Ann H.
Author_Institution :
Sci. Applications Int. Corp., Santa Barbara, CA, USA
Abstract :
This study was motivated by a need to simplify the operation of an existing multisensor perimeter intrusion-detection system, and by a step upward in processing power and storage capacity of small personal computers. The feasibility of centralizing and automating the processing was studied, and telemetry options were examined. While research is still needed to develop key algorithms, bandwidths and data rates have been checked for two processing structures and for perimeters in excess of 10 km and found to be easily within the capability of a centralized 386/1167 processing system
Keywords :
adaptive systems; electric sensing devices; microcomputer applications; safety systems; telemetering; adaptive real-time training strategies; bandwidths; centralized 386/1167 processing system; data rates; key algorithms; multisensor perimeter intrusion-detection system; processing power; processing structures; small personal computers; storage capacity; telemetry options; trainable multi-sensor perimeter intrusion detection; Data acquisition; Distributed processing; Event detection; Intrusion detection; Management training; Protection; Sensor fusion; Signal processing; System testing; Training data;
Conference_Titel :
Security Technology, 1988. Crime Countermeasures, Proceedings. Institute of Electrical and Electronics Engineers 1988 International Carnahan Conference on
Conference_Location :
Lexington, KY
DOI :
10.1109/CCST.1988.75996
