Sensor fusion for intelligent alarm analysis

Abstract

RECElV MAR 1 5 19%The purpose of an intelligent alarm analysis system is to provide complete and manageable information to a central alarm station operator by applying alarm processing and fusion techniques to sensor information. This paper discusses the sensor fusion approach taken to perform intelligent alarm analysis for the Advanced Exterior Sensor (AES). The AES is an intrusion detection and assessment system designed for wide-area coverage, quick deployment, low falsehuisance alarm operation, and immediate visual assessment. It combines three sensor technologies (visible, infixed, and millimeter wave radar) collocated on a compact and portable remote sensor module. The remote sensor module rotates at a rate of I revolution per second to detect and track motion and provide assessment in a continuous 360" field-of-regard. Sensor hsion techniques are used to correlate and integrate the track data &om these three sensors into a single track for operator observation. Additional inputs to the fusion process include environmental data, knowledge of sensor performance under certain weather conditions, sensor priority, and recent operator feedback. A confidence value is assigned to the track as a result of the fusion process. This helps to reduce nuisance alarms and to increase operator confidence in the system while reducing the workload of the operator. IntroductionA typical central alarm station for a security environment is an integrated system of people, procedures, and equipment. An alarm communication and display system receives alarm signals fiom intrusion detection sensors and displays the information to a security operator for action. Although annunciator displays observed by the operator are easy to understand and maintain, they typically contain only a limited amount of information. Therefore, the operator must manually assess each alarm occurrence, which could be difficult, time consuming, and error-prone. The idea behind intelligent alarm analysis (IAA) is to preprocess data from the security sensors and present alarm information to the security operator in a manner that would provide concise and meaningful information and increase confidence in true alarm events while filtering out the necessity to assess nuisance and false alarms. One of the goals of the IAA Project at Sandia National Laboratories is to devise this preprocessing methodology using alarm processing and fusion techniques on available sensor data.Initial intelligent alarm analysis concepts were developed for the Advanced Exterior Sensor (AES), an intrusion detection and assessment system currently being developed at Sandia. The AES requires the fusion of data

Locally constructed algorithms for distributed computations in ad-hoc networks

“…Fig. 3 shows the local implementation of the rule at the ith node corresponding to (20) in terms of the causal filterH(z) = z H(z), in the case that W is given by (12). Proper choice of the strictly causal filter h [k] in (20) with W in the form (12) can lead to AC rules with faster convergences rates than the associated first-order rules.…”

“…3 shows the local implementation of the rule at the ith node corresponding to (20) in terms of the causal filterH(z) = z H(z), in the case that W is given by (12). Proper choice of the strictly causal filter h [k] in (20) with W in the form (12) can lead to AC rules with faster convergences rates than the associated first-order rules. In particular, provided X k h[k] = 1, 1 We remark that we can readily construct AC rules for arbitrary connected topologies by adding to the denominator of the fraction on the right-hand side of (18) an > 0.…”

“…In particular, provided X k h[k] = 1, 1 We remark that we can readily construct AC rules for arbitrary connected topologies by adding to the denominator of the fraction on the right-hand side of (18) an > 0. Figure 3: Block diagram for admissible rule at the ith node corresponding to (20) with W given by (12) and whereH(z) = z H(z) is a causal filter.…”

“…To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. use in various civilian and military applications, including target tracking and surveillance for robot navigation [11,16] source localization [13] and radar applications [19], data gathering for weather forecasting and environmental applications [5,7,12], and medical monitoring and imaging [2,3,14]. In general, the networks envisioned for many of these applications involve large numbers of possibly randomly distributed inexpensive sensor nodes, with limited sensing, processing, and communication power on board.…”

Network traffic anomaly detection based on catastrophe theory

“…In these methods, the macro features of network traffic, such as self-similarity [1,2],entropy [3,4], probability distribution [5,6,7,8] et al, were extracted, followed by using various pattern recognition techniques, such as neural networks [9], hidden Markov model [10], integrated access control [11], and sensor fusion [12], machine learning [13] ,to detect network anomaly.…”