Optoelectronic InGaN Neuromorphic Synapse for Artificial Intelligence Applications

InGaN-based nanoscale light-emitting diodes (nanoLEDs) have emerged as promising optoelectronic synapses owing to reduced parasitic capacitance and shorter carrier lifetime. Their integration into neuromorphic systems holds immense promise for next-generation optical processing units in artificial vision, robotics, and edge computing—fields that demand real-time sensory processing and autonomous decision-making. In this study, we first demonstrate the synaptic behavior of InGaN nano-LEDs under various electrical pulse conditions as a proof-of-concept validation. Further, we extend our approach to characterize a <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$28\times 28$</tex> array of parallelly addressed nanoscale InGaN LEDs and compare their feasibility as high-density optical neuromorphic processing units.