The workflow of the automated extraction tool for simplified EKV parameters

Ultra-low-power time-domain neural accelerator achieves 116 TOPS/W using novel digital-to-time converter for IoT applications.

IEEE Xplore
Field: Electrical Standards
Electrical Standards
10Year<
Canada
Add us on Google

The increasing demand for high performance and energy efficiency in Artificial Neural Networks (ANNs) accelerators has driven a wide range of application-specific integrated circuits (ASICs). Besides, the rapid deployment of low-power IoT devices requires highly efficient computing, which as a result urges the need to explore low-power hardware implementations in different domains. This paper proposes a spatially unrolled time-domain accelerator that uses an ultra-low-power digital-to-time converter (DTC) while occupying an active area of 0.201 mm2. The proposed DTC is implemented using a Laddered, Inverter (LI) circuit, which consumes 3    ×    less power than the conventional inverter-based DTC and provides reliable performance across different process corners, supply voltages, and temperature variations. Post-synthesis results in 65nm CMOS show that the proposed core achieves a superior energy efficiency of 116 TOPS/W, a throughput of 4 GOPS, and an area efficiency of 20 GOPS/mm2. The proposed core improves energy efficiency by 2.4 - 47    ×    compared to prior time-domain accelerators.

Source: IEEE Xplore

Statement: Respect the original, good articles worth sharing, if there is infringement please contact delete.

Give a tip and encourage the author!

Comments

Share your question or perspective
WhatsApp
Send inquiry
+86
Click to upload file
Download
Get the IEE Business Application
Use the IEE-Business app to find equipment, obtain solutions, connect with experts, and participate in industry collaboration anytime, anywhere—fully supporting the development of your power projects and business.
Login
or continue with
New here?
Register