Speaker
Benjamin Parpillon
(Fermi National Accelerator Laboratory)
Description
We present a smart pixel prototype readout integrated circuit (ROIC), designed in a 28 nm CMOS bulk process, featuring in-pixel implementation of a machine learning based data filtering algorithm. This serves as a proof-of-principle for future particle tracking detectors. The ROIC integrates in-pixel signal processing with a fully connected two-layer neural network that analyzes information from a cluster of pixels. The network classifies, in real time, whether an incident particle has high or low momentum based on the pattern detected in the pixel array. In this talk, we will provide an overview of the chip architecture and present testing results from the first fabricated chip.
Authors
Anthony Badea
(University of Chicago)
Benjamin Parpillon
(Fermi National Accelerator Laboratory)