Mathematical models

A 3 nm FinFET 125 TOPS/W-29 TFLOPS/W, 90 TOPS/mm2-17 TFLOPS/mm2 SRAM-Based INT8, and FP16 Digital-CIM Compiler With Support for Multi-Weight Update/Cycle

A 3 nm FinFET 125 TOPS/W-29 TFLOPS/W, 90 TOPS/mm2-17 TFLOPS/mm2 SRAM-Based INT8, and FP16 Digital-CIM Compiler With Support for Multi-Weight Update/Cycle 150 150

Abstract:

This article presents an static random-access memory (SRAM)-based digital compute-in-memory (CIM) compiler implemented with 3 nm high- $\kappa $ metal gate (HKMG) FinFET technology, supporting flexible INT8 and FP16 formats for weight and activation multiply-accumulate (MAC) operations, offering configuration flexibility, high accuracy, and improved area and power efficiency. The FP16 digital …

View on IEEE Xplore

A Self-Injection LC Oscillator for Flicker Noise Reduction

A Self-Injection LC Oscillator for Flicker Noise Reduction 150 150

Abstract:

Self-injection has been used in lasers and photonic integrated circuits to reduce the laser’s phase noise (PN). We show that self-injection can be leveraged in GHz LC oscillators as well. Our oscillator employs a current-domain self-injection technique by leveraging second-harmonic extraction, capacitive phase shifting, and self-mixing through the oscillator’…

View on IEEE Xplore

Integrated Spatiotemporal Multiscale- Multiphysics-Uncertainty Simulation for Controlling Variability in RRAM Devices

Integrated Spatiotemporal Multiscale- Multiphysics-Uncertainty Simulation for Controlling Variability in RRAM Devices 150 150

Abstract:

Resistive random access memory (RRAM) is a leading candidate for next-generation nonvolatile memory and neuromorphic computing. However, its performance is limited by inherent switching variability and uncertainties in spatiotemporal multiscale materials and processes. This study integrates multiphysics and multiscale modeling with uncertainty quantification (UQ) to systematically address these limitations and …

View on IEEE Xplore

A Multicore Programmable Variable-Precision Near-Memory Accelerator for CNN and Transformer Models

A Multicore Programmable Variable-Precision Near-Memory Accelerator for CNN and Transformer Models 150 150

Abstract:

Convolutional neural network (CNN) and transformer are the most popular neural network models in computer vision (CV) and natural language processing (NLP). It is quite common to use both these two models in multimodal scenarios, such as text-to-image generation. However, these two models have very different memory mappings, dataflows and …

View on IEEE Xplore

A Compact, Highly-Digital Sensor-Fusion-Based Joint V dd-Temperature Sensor for SoC Thermal Management

A Compact, Highly-Digital Sensor-Fusion-Based Joint V dd-Temperature Sensor for SoC Thermal Management 150 150

Abstract:

This article presents a fine-grained thermal sensing network for thermal management in SoCs. Sensor nodes in this network are made up of joint supply voltage ( $V_{\mathrm {dd}}$ ) and temperature ( $T$ ) sensors, which are compact and highly digital. Measurements from these simple but imperfect sensors are jointly processed to extract …

View on IEEE Xplore

A SPICE-Compatible Compact Model of Ferroelectric Diode

A SPICE-Compatible Compact Model of Ferroelectric Diode 150 150

Abstract:

In this work, for the first time, we present a SPICE-compatible compact model of ferroelectric (FE) diodes to enable their design exploration for diverse applications, including memory and unconventional computing paradigms. We propose modified Schottky barrier and hopping models for capturing the on- and off-mode operations of the FE diode, …

View on IEEE Xplore

Integrating Atomistic Insights With Circuit Simulations via Transformer-Driven Symbolic Regression

Integrating Atomistic Insights With Circuit Simulations via Transformer-Driven Symbolic Regression 150 150

Abstract:

This article introduces a framework that establishes a cohesive link between the first principles-based simulations and circuit-level analyses using a machine learning-based compact modeling platform. Starting with atomistic simulations, the framework examines the microscopic details of material behavior, forming the foundation for later stages. The generated datasets, with molecular insights, …

View on IEEE Xplore

Quantum Field Theory Model for Spin-Based Devices Using 2-D van der Waals Materials

Quantum Field Theory Model for Spin-Based Devices Using 2-D van der Waals Materials 150 150

Abstract:

We explore the effects of layered geometries of 2-D quantum spin systems as a method to tune and control material properties for spintronic devices. We analyze the dispersion relation of a 2-D quantum spin system with a shifted bilayer square lattice through the linear spin wave (LSW) approximation of quantum …

View on IEEE Xplore

An 8-Bit 400-MS/s 1-Then-2-Bit/Cycle SAR ADC With Comparator Rotation-Based Input-Independent Background Offset Calibration

An 8-Bit 400-MS/s 1-Then-2-Bit/Cycle SAR ADC With Comparator Rotation-Based Input-Independent Background Offset Calibration 150 150

Abstract:

This letter presents an 8-bit 400 MS/s 1-then-2-bit/cycle successive approximation register (SAR) analog-to-digital converter (ADC) employing a comparator rotation-based background offset calibration (CRBC) technique. Unlike conventional 1-then-2-bit/cycle architectures, where calibration validity depends on the input voltage, the proposed comparator rotation-based background calibration enables input-independent background calibration, …

View on IEEE Xplore