Spiral Tip Recognition via Deterministic Learning

Abstract

The spiral tip is vital to the understanding of the spiral wave behaviors. Most studies of spiral tip dynamics focused on the prevention, control, and elimination of spiral wave, while few studies focused on the recognition of spiral wave. In real systems with the spiral wave, the recognition of the spiral wave should be before control or elimination. In the paper, we study the recognition of the spiral tip via deterministic learning. It mainly consists of two phases: the identification phase and the recognition phase. In the identification phase, the dynamics of spiral tips of the training set is accurately identified by using deterministic learning. In the recognition phase, a set of errors is obtained for a test spiral tip by employing an estimator model. Finally, the recognition of test spiral tip is achieved according to the smallest error principle. Numerical simulations based on the spiral tip generated by the Barkley model are performed to demonstrate the effectiveness and feasibility of the proposed method.