Various synchronization patterns from a pulse-coupled neural network of chaotic spiking oscillators

We present a pulse-coupled neural network (PCNN) of chaotic spiking oscillators (CSOs) with a spike-train input. The PCNN exhibits the following phenomena: each CSO synchronizes with a part of the spike-train input: 1) without overlapping or 2) with overlapping. We clarify parameter regions for these phenomena. The PCNN has various synchronization states and exhibits one of them depending on the initial state. The synchronization stability is caused by compulsory firings by the input. Using a simple test circuit, typical phenomena are verified in the laboratory