Week 07 - Neuromorphic Intelligence and Brain-Computer Interfaces

Topic Overview:

Introduce the core principles and architectural design of neuromorphic sensing and computing chips, as well as their applications in brain-computer interfaces (BCI). The lecture will focus on event-driven spiking neural networks (SNN), sense-compute integrated chips (e.g., Speck™), ultra-low-power analog front-ends (AFE), and high-throughput neural signal decoding algorithms. Participants will learn about the full hardware-software co-design pipeline from neural signal acquisition to intent decoding, and explore the industrial prospects of such technologies in scenarios including medical rehabilitation and intelligent interaction.

Key Topics:

1. Introduction of Research Institute for Artificial Intelligence of Things (RIAIoT)
2. Understand the basic principles of neuromorphic computing, event-driven architectures, and hardware mapping methods for spiking neural networks (SNN).
3. Master the underlying logic of the sense-compute integrated chip (Speck™ series) and the brain-computer interface signal chain (Rigi).
4. Analyse key technical challenges and solutions, including real-time decoding of thousand-channel neural signals and ultra-low-power analog front-ends.
   

Teaching Format

Lecture and case studies