Logical chaotic resonance in the FitzHugh-Nagumo neuron

It has been shown recently that a chaos-driven bistable system with two square waves as input can consistently work as a reliable logic gate, in an optimal window of chaos intensity. This phenomenon is referred to as logical chaotic resonance (LCR). In this study, the conventional LCR paradigm is extended to the excitable FitzHugh-Nagumo (FHN) neuron. Firstly, we have examined the intriguing possibility of achieving specific logic responses in the chaos-driven FHN neuron. Then, we have explained why LCR happens through plotting the phase portraits. Based on the Helmholtz's theorem, we have further discussed the relationship between success probability of obtaining specific logic operation, average spiking rate and average energy of the system. Lastly, for weak chaotic driving force, periodic force-assisted LCR can still be achieved, depending on appropriate amplitude and frequency of periodic driving force. Therefore, LCR can be controlled by regulating the frequency and amplitude of periodic driving force. Taken together, the chaos-driven FHN neuron can reliably function as logic gates even for subthreshold signals, thus consuming very low power. So the chaos-driven FHN neuron can potentially act as energy-efficient computing building block of futuristic neuromorphic systems with logic gate functionalities.