IEEE Journal of Solid-State Circuits – Early Access

A Scalable 1024-Channel Ultra-Low-Power Spike Sorting Chip With Event-Driven Detection and Spatial Clustering

A Scalable 1024-Channel Ultra-Low-Power Spike Sorting Chip With Event-Driven Detection and Spatial Clustering 150 150

Abstract:

This article presents a 1024-channel ultra-low-power spike sorting chip featuring event-driven spike detection and spatial clustering for large-scale neural recording. To address power and scalability constraints in brain–computer interfaces (BCIs), the design integrates a compressive analog-to-digital converter (ADC) with a two-stage spike detector that significantly reduces memory and processing …

View on IEEE Xplore

Ultra-Low-Power Dynamic-Bias Comparators With Self-Clocked Latch in 65-nm CMOS

Ultra-Low-Power Dynamic-Bias Comparators With Self-Clocked Latch in 65-nm CMOS 150 150

Abstract:

This article introduces two comparators featuring a dynamic-bias preamplifier and self-clocked latches, tailored for ultra-low-power and medium-speed applications with <500- $\mu $ V input-referred noise (IRN). The proposed self-clocked latches are activated by the preamplifier outputs and therefore operate with a lower common-mode current, which in turn minimizes the crowbar current …

View on IEEE Xplore

DPIM: A 2T1C eDRAM Transformer-in-Memory Chip With Sparsity-Aware Quantization and Heterogeneous Dense–Sparse Core

DPIM: A 2T1C eDRAM Transformer-in-Memory Chip With Sparsity-Aware Quantization and Heterogeneous Dense–Sparse Core 150 150

Abstract:

Transformer models have revolutionized artificial intelligence (AI) applications across various domains, but their increasing complexity poses significant challenges in terms of computational and memory demands. While processing-in-memory (PIM) paradigms have been adopted to address these limitations, existing PIM-based transformer accelerators still face hurdles such as: 1) focusing solely on optimizing attention …

View on IEEE Xplore

A 25.8% 3σ/μ-Accuracy, 0.12%/°C Temperature Drift Sigma-Delta Modulation Calibrated Pseudo-Resistor With GΩ to TΩ Tuning Range

A 25.8% 3σ/μ-Accuracy, 0.12%/°C Temperature Drift Sigma-Delta Modulation Calibrated Pseudo-Resistor With GΩ to TΩ Tuning Range 150 150

Abstract:

This article presents an on-chip self-calibrated pseudo-resistor (PR) based on a sigma-delta modulation (SDM) loop. The proposed real-time calibration mechanism facilitates the implementation of an accurate low-drift ultra-high-value (UHV) PR with a wide tuning range at minimum hardware expenditure. Experimental results demonstrate that the resistance can be precisely tuned from 1.5 …

View on IEEE Xplore

A Low-Jitter Fractional-N Digital PLL Using a Quantization-Error-Compensating BBPD and an Orthogonal-Polynomial-Based LMS Calibration

A Low-Jitter Fractional-N Digital PLL Using a Quantization-Error-Compensating BBPD and an Orthogonal-Polynomial-Based LMS Calibration 150 150

Abstract:

This work presents a 10.0–11.5-GHz fractional- $N$ digital phase-locked loop (DPLL) using the quantization-error-compensating bang–bang phase detector (QEC-BBPD) that can minimize both the static delay ( $T_{\mathrm {S}}$ ) and the dynamic delay ( $T_{\mathrm {D}}$ ) required for removing the delta-sigma modulator’s ( $\Delta \Sigma $ M) quantization-error (Q-error). Since the …

View on IEEE Xplore

A 0.5-/0.95-dB NF, 50-/25-Ω Configurable CMOS Front-End ASIC for the Readout of Liquid Argon Calorimeter in the LHC

A 0.5-/0.95-dB NF, 50-/25-Ω Configurable CMOS Front-End ASIC for the Readout of Liquid Argon Calorimeter in the LHC 150 150

Abstract:

This article presents the design of a four-channel front-end application specific integrated circuit (ASIC), ATLAS liquid argon front-end (ALFE), developed for the readout of the liquid-argon calorimeter (LAr) detector in the ATLAS experiment at the Large Hadron Collider (LHC). ALFE enables the readout of current signals induced in the LAr …

View on IEEE Xplore

A 180-nm Voltage-Controlled Magneto-Electric RAM With Sub-1-ns Switching Time

A 180-nm Voltage-Controlled Magneto-Electric RAM With Sub-1-ns Switching Time 150 150

Abstract:

Memory performance has emerged as a critical factor influencing both system speed and energy efficiency. However, conventional memory technologies such as embedded Flash (eFlash) and static RAM (SRAM) encounter significant scalability limitations beyond the 28-nm CMOS node. Among novel emerging memory technologies, spin-transfer-torque magnetic RAM (STT-MRAM) has gained prominence due …

View on IEEE Xplore

EdgeDiff: Energy-Efficient Multi-Modal Few-Step Diffusion Model Accelerator Using Mixed-Precision and Reordered Group Quantization

EdgeDiff: Energy-Efficient Multi-Modal Few-Step Diffusion Model Accelerator Using Mixed-Precision and Reordered Group Quantization 150 150

Abstract:

Recent advances in diffusion models (DMs)—such as few-step denoising and multi-modal conditioning—have significantly improved computational efficiency and functional flexibility, but they also introduce new hardware challenges. In particular, the elimination of inter-timestep redundancy, increased encoder/decoder workload, and heightened sensitivity to quantization demand a new class of accelerator. …

View on IEEE Xplore

A Radiated-EMI-Reduced Touch AFE IC With Pipelined Dual-Frequency Modulation and Sine2 Waveform Shaping for Automotive Applications

A Radiated-EMI-Reduced Touch AFE IC With Pipelined Dual-Frequency Modulation and Sine2 Waveform Shaping for Automotive Applications 150 150

Abstract:

This article presents an analog front-end (AFE) for automotive touch-screen sensors that reduces radiated emission (RE) from touch-screen panel (TSP) driving signals while maintaining touch sensing performance. The proposed touch AFE employs a sine–sine (Sine2) waveform shaping and a sine-chirp spread spectrum (SCSS) clocking to suppress both harmonic components …

View on IEEE Xplore