[2023] Porting a neural-recording kernel on FPGA & Memristor-based circuits for next-generation Brain-Machine Interfaces

One popular approach for understanding brain function is by studying electrophysiological recordings from animals. One class of neural recording involves acquiring Local-Field potentials (LFPs), which are transient electrical signals generated in nervous and other tissues by the summed and synchronous electrical activity of the neurons in that tissue. LFPs are typically recorded with a high-impedance microelectrode placed in the midst of the population of cells generating it. For high-resolution recordings (i.e., the ones using multi-electrode arrays or MEAs), the volume of the recorded neural data becomes significant, which makes it difficult to transmit it wirelessly to the outside world. Thus, neuroscientists currently rely on wired setups to transmit this data, which limit the free movement of the animals. To solve this problem, we recently developed a kernel that performs dimensionality reduction on the sparse LFP data in order to efficiently transfer this information to an external setup for offline analysis.

Assignment

The goal of this thesis is to port the LFP post-processing kernel to an FPGA and a memristor-based crossbar (ReRAM) in order to achieve small-form-factor dimensionality reduction for brain-machine interfaces (BMIs).

Requirements

MSc student in Computer Engineering, Embedded Systems or Biomedical Engineering with a profile in signal processing (basic), hardware design and, optionally, Python.

Contact

Last modified: 2022-10-04