A 65-nm Privacy-Preserving Neuromorphic Encoder With 7.13-nJ Efficiency, 2.38-Mb/mm^2 Item-Memory Density, and Federated Learning Support

2026-06-05

AI Summary

This work addresses the dual demands of energy efficiency and privacy in intelligent assistants and wearable devices by proposing a 65 nm neuromorphic encoder that integrates physical unclonable functions (PUFs) with hyperdimensional computing (HDC). The design features an innovative 2T-2T entropy cell that leverages transistor process variations to generate device-specific, write-free item memory, eliminating the need for storing random basis vectors and thereby reducing hypervector dimensionality by 14.3×. The chip supports in-situ inference, continual learning, and multi-user federated learning. Evaluated on EMG and UCI-HAR datasets, it achieves classification accuracies of 93.2% and 96.1%, respectively, with an encoding energy of only 7.13 nJ, an item memory density of 2.38 Mb/mm², and training and inference energies of 357.32 nJ and 76.44 nJ, respectively.

Abstract

The increasing demand for privacy-preserving personal data analytics in smart assistants, wearable health monitors, and context-aware systems calls for hardware that is both energy-efficient and secure. This work presents a 65-nm privacy-preserving neuromorphic encoder that leverages transistor-level process variation as physically unclonable entropy for hyperdimensional computing. The proposed 2T-2T entropy cell enables compact, device-specific, and write-free item memory, allowing privacy-preserving bio-signal encoding without storing random basis vectors in conventional memory. The fabricated prototype achieves 7.13 nJ per encoding, 2.38 Mb/mm^2 item-memory density, 76.44 nJ per prediction, and 357.32 nJ per training update. It also supports in-situ decision-making, continual learning, and federated learning for multi-user deployment and cold-start personalization. Evaluations across bio-signal datasets demonstrate 93.2% accuracy on EMG and 96.1% accuracy on UCI-HAR, while reducing hypervector dimensionality by 14.3x compared with binary hyperdimensional computing. These results demonstrate an energy-efficient and privacy-preserving neuromorphic hardware platform for secure edge biomedical intelligence.

Problem

• privacy-preserving
• neuromorphic computing
• edge intelligence
• bio-signal encoding
• federated learning

Innovation

• neuromorphic encoder
• physically unclonable entropy
• hyperdimensional computing
• federated learning
• privacy-preserving hardware