The detection of transient‐like signals in the presence of a spatially inhomogeneous, nonstationary noise field
Manuel T. Silvia · International Journal of Imaging Systems and Technology · 1989
Abstract We will consider the problem of detecting transient‐like signals by means of a spatially distributed array of sensors embedded in an inhomogeneous, nonstationary noise field. If each element of the array has an independent, uncorrupted reference sensor to estimate the noise statistics, then conventional adaptive noise cancellation algorithms can be used to improve the detection process. However, because of practical real‐world constraints, independent, uncorrupted reference sensors for the noise might not be available. Thus, the applicability of conventional adaptive noise cancellation techniques is in question. This paper will discuss the development of a knowledge‐based signal‐processing system that uses Artificial Intelligence (Al) methodologies to adaptively cancel inhomogeneous, nonstationary noise from a distributed array of passive sensors that is constrained to have no noise‐reference sensors.