Circuits

PERCEL: A Rewritable NVM CIM Incorporating a CTT-Based Per-Cell DAC

PERCEL: A Rewritable NVM CIM Incorporating a CTT-Based Per-Cell DAC 150 150

Abstract:

Compute-in-memory (CiM) accelerators perform matrix vector multiplications (MVMs) directly inside memory arrays, reducing data movement and improving both energy efficiency and throughput for artificial intelligence (AI) workloads. To reduce the number of conversions, recent designs use multibit compute cells. Nevertheless, practical multibit CiM still faces a tension among accuracy, efficiency, …

View on IEEE Xplore

A Reconfigurable Multimode Readout IC for Nonconductive and Capacitive Sensing With 33.9-dB SNR on 6.5-in AMOLED Panels

A Reconfigurable Multimode Readout IC for Nonconductive and Capacitive Sensing With 33.9-dB SNR on 6.5-in AMOLED Panels 150 150

Abstract:

This letter presents a multimode readout IC for 6.5-in 34Tx/16Rx on-cell touch AMOLED (OCTA) panels, detecting both conductive and nonconductive objects (NCos) without panel modifications. To enable this dual detection, a reconfigurable analog front-end (AFE) is proposed, functioning as either a capacitance-to-voltage converter or a triboelectric charge sampler. In …

View on IEEE Xplore

OISMA: On-the-Fly In-Memory Stochastic Multiplication Architecture for Approximate Matrix Multiplication

OISMA: On-the-Fly In-Memory Stochastic Multiplication Architecture for Approximate Matrix Multiplication 150 150

Abstract:

Artificial intelligence (AI) models are currently driven by a significant upscaling of their complexity, with massive matrix-multiplication workloads representing the major computational bottleneck. In-memory computing (IMC) architectures are proposed to avoid the von Neumann bottleneck. However, both digital/binary-based and analog IMC architectures suffer from various limitations, which significantly degrade …

View on IEEE Xplore

A Novel VGSOT-pMTJ Write Circuit for Hybrid CMOS/MTJ CIM Architecture

A Novel VGSOT-pMTJ Write Circuit for Hybrid CMOS/MTJ CIM Architecture 150 150

Abstract:

Hybrid computation-in-memory (CIM) architecture has emerged as the most promising alternative to overcome the drawbacks of the conventional CMOS-only devices used in the conventional von-Neumann architecture. In the hybrid CIM architecture, a pair of perpendicular magnetic tunnel junctions (pMTJs) is used to store one bit of information. Though there are …

View on IEEE Xplore

A High-Efficiency Magnetoelectric Power Link With a Load-Adaptive CMOS Rectifier for Miniaturized Implants

A High-Efficiency Magnetoelectric Power Link With a Load-Adaptive CMOS Rectifier for Miniaturized Implants 150 150

Abstract:

Miniaturized biomedical implants are currently constrained by limited wireless power transfer (WPT) efficiency under load variation and spatial misalignment. This work presents a custom magnetoelectric (ME) power link incorporating a load-adaptive CMOS rectifier to address these challenges. By employing a current sensing control loop for dynamic switch sizing and delay …

View on IEEE Xplore

LUT-Based Convolutional Tsetlin Machine Accelerator With Dynamic Clause Scaling for Resources-Constrained FPGAs

LUT-Based Convolutional Tsetlin Machine Accelerator With Dynamic Clause Scaling for Resources-Constrained FPGAs 150 150

Abstract:

The rapid growth of machine learning (ML) workloads, particularly in computer vision applications, has significantly increased computational and energy demands in modern electronic systems, motivating the use of hardware accelerators to offload processing from general-purpose processors. Despite advances in computationally efficient ML models, achieving energy-efficient inference on resource-constrained edge devices …

View on IEEE Xplore

Characterization and Modeling of Multilevel Analog ReRAM Synapses in the Sky130 Process

Characterization and Modeling of Multilevel Analog ReRAM Synapses in the Sky130 Process 150 150

Abstract:

Nonvolatile memory devices play a key role in enabling energy-efficient computing. Among them, analog nonvolatile memories such as resistive random access memory (ReRAM) offer high density and low power compared to conventional digital memories. However, their analog nature introduces device-level variability that impacts computational accuracy. This work presents the characterization …

View on IEEE Xplore

A 142–164-GHz Phased-Array AiP Module With High-Power-Density and High-Efficiency Transceiver in 65-nm CMOS for 6G UE

A 142–164-GHz Phased-Array AiP Module With High-Power-Density and High-Efficiency Transceiver in 65-nm CMOS for 6G UE 150 150

Abstract:

This work presents a D-band high-power-density four-element phased-array transceiver for 6G user equipment (UE). Conventional designs require large multi-stage LO generation circuits for D-band up/down conversion, making it difficult to achieve compact size and low-power consumption. To address this, we propose an integrated LO chain using an injection-locked tripling …

View on IEEE Xplore

1–1.7-GHz Single-Voltage-Controlled Tunable MMIC BPF With Over 45-dB Rejection to 30 GHz Using Tapped Inductor Technique

1–1.7-GHz Single-Voltage-Controlled Tunable MMIC BPF With Over 45-dB Rejection to 30 GHz Using Tapped Inductor Technique 150 150

Abstract:

This article presents a novel fourth-order single-voltage-controlled (SVC) tunable monolithic microwave integrated circuit (MMIC) bandpass filter (BPF) for tunable filter bank applications. Utilizing the tapped inductor technique (TIT), the proposed filter achieves compact magnetic coupling with reduced loss, avoiding the use of bulky inductor- or transformer-based coupling structures. The filter …

View on IEEE Xplore