Distributed Detection of Signal Waveforms in Additive Gaussian Observation Noise.

摘要

This paper is concerned with the detection of signal waveforms by a distributed surveillance network comprised of: a collection of spatially separated sensors, and local signal processors collocated with the sensors. The local signal processors are assumed to implement likelihood ratio tests to detect the presence or absence of the signals. Signal detections may be used for local decisionmaking or passed upward to a fusion center for further processing. In either case, the local detection thresholds cannot be determined independently, but must be determined jointly to optimize overall surveillance system performance. Results are presented concerning the nature of this threshold computation for a number of interesting cases. (Author)