Optimum object classification with active sonar: new theoretical results

The active sonar classification problem is considered. To perform active classification, a known waveform is transmitted into a medium and directed towards a region called the test volume. An array of sensors is used to pick up the backscattered signal energy reflected from the cells of the test volume. The data are input into a processing algorithm. The processor is to decide whether an object is present and, if so, what kind of object is present. A systematic method for evaluating classifier performance is derived. A sensitivity analysis and interpretation of processor performance is made. An analysis of processor performance as a function of angular resolution is described