Hardware Implementation of Ring Oscillator Networks Coupled by BEOL Integrated ReRAM for Associative Memory Tasks

2025-03-18

AI Summary

To address the low coupling density and poor energy efficiency of oscillatory neural networks (ONNs) in hardware implementations, this work proposes an in-memory coupling architecture based on a back-end-of-line (BEOL)-integrated conductive metal oxide/HfOₓ resistive random-access memory (ReRAM) crossbar array. We present the first hardware ONN prototype comprising ring-oscillator-based neurons tightly coupled with a dense ReRAM crossbar, enabling in-situ processing and associative retrieval of phase-encoded analog signals. A novel brain-inspired associative memory paradigm—phase-locking–driven synchronization—is introduced and experimentally demonstrated on a 2×2 network for binary phase-pattern retrieval. Fabrication and characterization confirm the architectural feasibility, scalability, and high-density coupling capability at the hardware level. This approach establishes a promising pathway toward ultra-low-power neuromorphic computing.

Technology Category

• Machine Learning: Hardware-aware ML
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Economics, Online Markets and Human Computation: Social networks and social learning
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

We demonstrate the first hardware implementation of an oscillatory neural network (ONN) utilizing resistive memory (ReRAM) for coupling elements. A ReRAM crossbar array chip, integrated into the Back End of Line (BEOL) of CMOS technology, is leveraged to establish dense coupling elements between oscillator neurons, allowing phase-encoded analog information to be processed in-memory. We also realize an ONN architecture design with the coupling ReRAM array. To validate the architecture experimentally, we present a conductive metal oxide (CMO)/HfOx ReRAM array chip integrated with a 2-by-2 ring oscillator-based network. The system successfully retrieves patterns through correct binary phase locking. This proof of concept underscores the potential of ReRAM technology for large-scale, integrated ONNs.

Problem

• Hardware implementation of oscillatory neural networks using ReRAM.
• Integration of ReRAM in CMOS for in-memory analog information processing.
• Validation of ONN architecture with phase locking for pattern retrieval.

Innovation

• ReRAM crossbar array for dense coupling
• Phase-encoded analog in-memory processing
• CMO/HfOx ReRAM integrated with oscillators