# Resonant Energy Recycling SRAM Architecture

arXiv: https://arxiv.org/abs/2010.01767 · HTML: https://mlsanity.com/paper/2010.01767/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Oct 2020.

## Built on by (most cited first, 5 of 5)

- [Descriptor: Face Detection Dataset for Programmable Threshold-Based Sparse-Vision](https://mlsanity.com/paper/2410.00368/) (2024)
- [PGR-DRC: Pre-Global Routing DRC Violation Prediction Using Unsupervised Learning](https://mlsanity.com/paper/2507.13355/) (2025)
- [Time--to--Digital Converter (TDC)--Based Resonant Compute--in--Memory for INT8 CNNs with Layer--Optimized SRAM Mapping](https://mlsanity.com/paper/2601.00434/) (2026)
- [LiteEvent-AE: Lightweight Autoencoder for Event-Based Vision on Low-Latency Energy-Constrained Edge Devices](https://mlsanity.com/paper/2608.21764/) (2026)
- [DVA-Neurons: Design and Verification of Adaptive LIF Neurons: From Single-Neuron Dynamics to Multi-Neuron Spiking Networks](https://mlsanity.com/paper/2609.22775/) (2026)

## Abstract (arXiv)

Although we may be at the end of Moore's law, lowering chip power consumption is still the primary driving force for the designers. To enable low-power operation, we propose a resonant energy recovery static random access memory (SRAM). We propose the first series resonance scheme to reduce the dynamic power consumption of the SRAM operation. Besides, we identified the requirement of supply boosting of the write buffers for proper resonant operation. We evaluated the resonant 144KB SRAM cache through SPICE and test chip using a commercial 28nm CMOS technology. The experimental results show that the resonant SRAM can save up to 30% dynamic power at 1GHz operating frequency compared to the state-of-the-art design.