We present a novel angular fingerprinting algorithm for detecting changes in
the direction of rotation of a target with a monostatic, stationary sonar
platform. Unlike other approaches, we assume that the target's centroid is
stationary, and exploit doppler multipath signals to resolve the otherwise
unavoidable ambiguities that arise. Since the algorithm is based on an
underlying differential topological theory, it is highly robust to distortions
in the collected data.