Make the Signal Chain More Intelligent and Efficient with Mixed Signal Processing and In-Memory Computing

Explore a conference talk from tinyML Asia 2021 that delves into making signal chains more intelligent and efficient through mixed signal processing and in-memory computing. Learn about Reexen's innovative architecture that breaks down signal processing into mixed signal low-level feature extraction before digitization and mixed signal high-level in-memory computing after digitization. Discover how this approach offers significant improvements in energy consumption and cost efficiency compared to traditional signal chains. Gain insights into neuromorphic analog signal processing, advantages of analog signal processing, analog AI curves, and in-memory computing. Understand the challenges of traditional signal chains, including AD conversion bottlenecks and power consumption issues related to data transfer. Examine use cases for analog AI and digital architecture implementations. The talk concludes with audience questions and acknowledgment of sponsors.

Introduction
About the company
Research Background
Neuromorphic Analog Signal Processing
Traditional Analog Signal Processing
Advantages of Analog Signal Processing
Analog AI Curves
Analog Computing
Memory Computing
Energy Cost
InMemory Computing
OnetoOne Mapping
Efficiency
Conclusion
Analog AI use case
Digital architecture
Audience questions
Sponsors