New Methods for Nonlinear Tracking and Nonlinear Chaotic Signal Processing

We propose a method to identify, separate, and classify complex broad-band signals, and, subsequently, to track and locaiize the typically elusive threats emitting those signals. The proposed methods were developed by the authors, and have already undergone some basic preliminary testing. The classification portion of the algorithm allows one to separate, classify, and identify targets when the received signal is the sum of a few nonlinearly generated components in additive noise. It is therefore particularly relevant to realistic scenarios, unlike other proposed algorithms that cannot handle signals with multiple nonlinear components and/or additive noise. The tracking portion of the algorithm allows one to detect and follow a maneuvering target that is emitting either a nonlinearly generated signal, or a near-linear narrow-band continuous wave (CW), even when the signal is too faint to be seen in a single, typically short, time window.