Analysis and Optimization of Pixel Noise in Dynamic Vision Sensors

Abstract

To date, pixel noise in a dynamic vision sensor (DVS) has not been accurately analyzed in the literatures, and optimization of pixel noise has been performed empirically. This dissertation presents a theoretical analysis of the DVS pixel noise. The mean squared noise voltage at the pixel output from each noise source in a pixel is mathematically derived and verified through simulations and measurements. A pixel design method to achieve target noise performance while maintaining pixel latency is also presented based on the noise analysis. A prototype DVS chip was fabricated in a 110nm CMOS image sensor process and tested under various light and pixel bias conditions. It is shown that the proposed noise analysis and design method successfully predict the noise performance of the DVS chip.