Neuromorphic chip interface with a microcontroller - Master project

alternate text

The Institute of Neuroinformatics (INI) is a joint institute of the University of Zurich and the Swiss Federal Institute of Zurich (ETH Zurich). INI carries out experimental, theoretical and applied research with the aim of discovering key principles by which the brain is built and works, and using this knowledge in practical applications where possible. The Neuromorphic Cognitive Systems group at INI has been developing neuromorphic processing platforms that enable event-driven computing for sensory-processing at ultra-low power. These technologies perfectly fit the global trend towards dedicated hardware computing solutions for edge-computing applications that cannot connect to the cloud.

Up to now, the readout of the chips has been done by an external FPGA, which is very efficient timewise, but represents a big cost - and power - overload for our projects. With the recent advances in microcontroller technologies, we would like to explore the feasibility of the readout by a small microcontroller. Such an approach would reduce costs and development time, and potentially allow light on-chip pre-processing.

This project will allow a highly motivated student to explore input/output solutions of spikes generated by (or sent to) neuromorphic chips using the Teensy 4.0 platform (ARM Cortex M7 @ 600MHz). The student will get access to state-of-the-art neuromorphic computing platforms to benchmark the developed architecture (bandwidth, jitter) and to evaluate the temporal (1us resolution) and bandwidth (USB3 link to computer) requirements.

Requirements

Microcontroller theory, programming in C++, python, PCB design, basics of VHDL/Verilog.

Contact

Germain Haessig: germain (at) ini.uzh.ch
Giacomo Indiveri: giacomo (at) ini.uzh.ch