A comparison of algorithms for anomaly detection in safeguards and computer security systems using neural networks

Conference · Wed Jan 01 00:00:00 EST 1992

OSTI ID:7117430

Howell, J A; Whiteson, R

Detection of anomalies in nuclear safeguards and computer security systems is a tedious and time-consuming task. It typically requires the examination of large amounts of data for unusual patterns of activity. Neural networks provide a flexible pattern-recognition capability that can easily be adapted for these purposes. In this paper, we discuss architectures for accomplishing this task.

• Research Organization: Los Alamos National Lab., NM (United States)
• Sponsoring Organization: USDOE; USDOE, Washington, DC (United States)
• DOE Contract Number: W-7405-ENG-36
• OSTI ID: 7117430
• Report Number(s): LA-UR-92-2214; CONF-9207102-27; ON: DE92018637
• Resource Relation: Conference: Institute of Nuclear Materials Management (INMM) annual meeting, Orlando, FL (United States), 19-22 Jul 1992
• Country of Publication: United States
• Language: English

Related Subjects

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL PROTECTION
COMPUTERIZED CONTROL SYSTEMS
NEURAL NETWORKS
NUCLEAR FACILITIES
NUCLEAR MATERIALS MANAGEMENT
SECURITY
ALGORITHMS
SAFEGUARDS
CONTROL SYSTEMS
MANAGEMENT
MATHEMATICAL LOGIC
ON-LINE CONTROL SYSTEMS
ON-LINE SYSTEMS
055001* - Nuclear Fuels- Safeguards
Inspection
& Accountability- Technical Aspects