Sensing, tracking, and reasoning with relations - IEEE Signal Processing Magazine

Suppose we have a set of sensor nodes spread over a geographical area. Assume that these nodes are able to perform processing as well as sensing and are additionally capable of communicating with each other by means of a wireless network. The sensors may include small video or still cameras, acoustic microphone arrays, passive infrared intrusion (PIR), seismic, or magnetic sensing devices. Examples of such integrated devices include the UCLA/ Sensoria WINS nodes [33] or the Berkeley motes [27]. The sensors may be controllable in limited ways: they can activated, possibly aimed, commanded to sense and transmit data to their neighbors, etc. Some may be even mounted on mobile ground or air platforms and directed to move to specific locations. Though each node is an independent hardware device, they need to coordinate their sensing, computation, and communication to acquire relevant information about their environment so as to accomplish some high-level task. The integration of processing makes such nodes more autonomous and the entire system, which we call a sensor net, becomes a novel type of sensing, processing, and communication engine. Such sensor nets can be deployed in a variety of civilian and military contexts, including factory automation, environmental monitoring, health-care delivery, security and surveillance, and battlefield awareness. Their advantages over alternate architectures have been argued, for example, in [33]. Briefly, these include: Deployment can be rapid, as no cabling infrastructure is needed; Sensor placement close to signal sources results in better signal-to-noise (SNR) ratio; On-board processing allows the sensors to communicate compressed signal data; The system can be made robust against the failure of a small number of its sensor nodes;