Unattended monitoring of suspicious behavior for route surveillance

Abstract

A priori information on suspicious behaviour is extremely valuable for countering threats involving improvised explosive devices (IEDs). Suspicious activities along routes during expeditionary operations can be monitored by unattended networks using simple sensing nodes that can gather data for continuous monitoring of daily vehicle activity. Dedicated software yields the necessary intelligence on these activities by filtering suspicious behaviour from anomalous behaviour (including false alarms). Research has started to equip a commercially available sensor network with data analysis software. It aims at demonstrating the detection of suspicious behaviour along roads, within a required time span. Three phases are distinguished. First phase is the analysis of traffic flux in a simple scenario with three networks lying at three junctions. The second phase investigates the ability to track and classify one object in this scenario, while the third phase aims to track and classify two or more objects. Findings are presented for phase one, flux measurements.A priori information on suspicious behaviour is extremely valuable for countering threats involving improvised explosive devices (IEDs). Suspicious activities along routes during expeditionary operations can be monitored by unattended networks using simple sensing nodes that can gather data for continuous monitoring of daily vehicle activity. Dedicated software yields the necessary intelligence on these activities by filtering suspicious behaviour from anomalous behaviour (including false alarms). Research has started to equip a commercially available sensor network with data analysis software. It aims at demonstrating the detection of suspicious behaviour along roads, within a required time span. Three phases are distinguished. First phase is the analysis of traffic flux in a simple scenario with three networks lying at three junctions. The second phase investigates the ability to track and classify one object in this scenario, while the third phase aims to track and classify two or more objects. Findings are presented for phase one, flux measurements.