ESTIMATION AND DETECTION ISSUES IN MATCHED-FIELD PROCESSING

The primary purpose of this paper is to establish a conceptual framewo rk in which the model-based, space-time acoustic signal processing pro cedure known as matched. field processing (MFP) can be handled in a co nsistent manner. After a discussion of the history and background of t he subject, a framework for strong-signal MFP is developed, based on s tandard statistical estimation therapy. The point of view is taken tha t MFP is, at its core, an estimation problem in the strong-signal regi me. In the weak-signal cases, the necessary requirement of detection d ictates that MFP then be considered a joint detection-estimation task. It is demonstrated that, generally, MFP is essentially a space-time p rocessing problem rather than simply an array processing (spatial proc essing only) procedure. Following this framework, an overview of the p rocessing schemes used to date in MFP is given in which it is shown ho w these methods relate to the optimal space-time structure. Weak-signa l detection and estimation scenarios relevant to MFP are then noted. A discussion of present methods for dealing with the inherent instabili ty of MFP algorithms (mismatch) is also presented. The final section c ontains a concise critical summary of the current status of MFP, with some general recommendations for future research.