Criticality and Neuromorphic Sensing in a Single Memristor

Resistive random access memory (RRAM) is an importanttechnologyfor both data storage and neuromorphic computation, where the dynamicsof nanoscale conductive filaments lies at the core of the technology.Here, we analyze the current noise of various silicon-based memristorsthat involves the creation of a percolation path at the intermediatephase of filament growth. Remarkably, we find that these atomic switchingevents follow scale-free avalanche dynamics with exponents satisfyingthe criteria for criticality. We further prove that the switchingdynamics are universal and show little dependence on device sizesor material features. Utilizing criticality in memristors, we simulatethe functionality of hair cells in auditory sensory systems by observingthe frequency selectivity of input stimuli with tunable characteristicfrequency. We further demonstrate a single-memristor-based sensingprimitive for representation of input stimuli that exceeds the theoreticallimits dictated by the Nyquist-Shannon theorem.