Excitability and synchronization of vanadium dioxide memristor-inspired neurons

Neuromorphic devices play a significant role in exploiting the dynamical analogy between various physical circuits and neuronal systems. Mott memristive systems made from vanadium dioxide (VO₂) are candidates for neuromorphic computations because of their ability to better reproduce neuron-like functions/behaviors and operate at low transition energy. In this paper, we revisit a prior work on the VO₂ memristor-inspired neuron to not only identify the exact biophysical mechanisms of this system's features, but also reproduce the more extensive neuron-like dynamical behaviors as two types of excitability and spiking by combining one-parameter bifurcations with two-parameter panels. Based on the Lyapunov stability theorem, a novel criterion for exponential synchronization is acquired in coupled VO₂ memristor-inspired neurons by introducing two controllers. Meanwhile, it is demonstrated via numerical simulations and hardware circuits. Results provide the intersection of electronic physics and theoretical neuroscience from the nonlinear dynamics point of view.