ASAP: A Camera Sensor Network for Situation Awareness

Abstract

Abstract.Situation awareness is an important application category in cyber-physical systems, and distributed video-based surveillance is a good canonical example of this application class. Such applications are interactive, dynamic, stream-based, computationally demanding, and needing real-time or near real-time guarantees. A sense-process-actuate control loop characterizes the behavior of this application class. ASAP is a scalable distributed architecture for a multi-modal sensor network that caters to the needs of this application class. Features of this architecture include (a) generation of prioritization cues that allow the infrastructure to pay selective attention to data streams of interest; (b) virtual sensor abstraction that allows easy integration of multi-modal sensing capabilities; and (c) dynamic redirection of sensor sources to distributed resources to deal with sudden burstiness in the application. In both empirical and emulated experiments, ASAP shows that it scales up to a thousand of sensor nodes (comprised of high bandwidth cameras and low bandwidth RFID readers), significantly mitigates infrastructure and cognitive overload, and reduces false negatives and false positives due to its ability to integrate multi-modal sensing.

Target container: A target-centric parallel programming abstraction for video-based surveillance

“…Other projects that are related to TC include ASAP [10]; the activity topology design based surveillance middleware [11]; service-oriented architectures for sensor networks such as OASIS [12]; and the high level abstractions for sensor network programming [13], [14]. ASAP [10] provides scalable resource management by using application-specific prioritization cues.…”

“…ASAP [10] provides scalable resource management by using application-specific prioritization cues. Hengel et al [11] approach scalability by partitioning the system according to an activity topology describing the observed (past) behavior of target objects in the network.…”

Persistent Temporal Streams

“…The compressed streams' latencies are affected too (but not as severely) because the same node and disk are used to host both types of streams. To provide a frame of reference for our numbers, ASAP [4] is a video surveillance system implemented in Java; its published end-to-end latency results for live queries are between 135-175 ms, which in practice are perfectly adequate for the application domain. In our PTS-based evaluation, the highest latency we have measured for historical queries is 18ms (+/-16ms).…”

“…-A persistent temporal stream abstraction targeted for live stream analysis applications (Section 2) -An architecture for stream persistence and an analysis of the potential design space of persistent temporal streams (Sections 2.1 & 2.2) -A system design and implementation of PTS, a distributed runtime realizing the persistent temporal stream abstraction (Section 3) -A system-level experimental evaluation of PTS and an application-based evaluation using a video-based surveillance system [4] (Section 4)…”

Tweeting Cameras for Event Detection

“…From visual sensors, wearable sensors, to humans as sensors [31], multiple sensor media streams constantly provide observations on the real world that are utilized in many aspects of our life: from industrial process control, robotics, surveillance, smart houses to situation awareness [34]. Examples of sensors include (a) physical sensors like cameras, accelerometers, gyroscopes, mobile phones, RFID tags, temperature sensors, humidity sensors and (b) social sensors like social networking sites containing user-generated content reporting events in all kinds of formats (text, image, video, etc.).…”