ISSCC 2026
Session 36
AI / ML
A Sparsity-Aware Neural Interface with CIM-Based Predictive Focused Sampling for Hotspot Spike Tracking
Chinese Academy of Sciences, Beijing, China, 4Soochow University, Suzhou, China *Equally Credited Authors (ECAs) 1 3 Abstract This 1024-electrode neural interface solves the power, area, and bandwidth bottleneck by combi
ISSCC 2026
Session 36
AI / ML
A 90.7%-Efficiency Hybrid Optogenetic Stimulation System with Sub-Threshold Electrical Stimuli Achieving 70.6% Optical Energy Saving
Abstract This work presents a hybrid optogenetic stimulation system with sub-threshold electrical stimuli that lower the optical threshold and reduce energy consumption. It achieves 90.75% LED driving efficiency through
ISSCC 2026
Session 36
AI / ML
An 80×80μm2/Pixel 55.48dB-Wide-DR 400-Pixel Subretinal Prosthesis SoC with Power-Aware Light Adaptation and Charge-Recycling Local Dynamic Supply
Abstract We propose a light-adaptive retinal prosthesis (LARP) SoC, which employs power-aware light adaptation achieving a wide light-sensing DR of 55.48dB. Pixel clustering enables a compact pixel size of 80×80μm2, enha
ISSCC 2026
Session 36
AI / ML
A CMOS Neural Probe with 1280 Electrodes and 88 Emission Sites Featuring Thermo-Optic Switching and On-Chip Calibration for Dual-Wavelength Optogenetics
Harrie A. C. Tilmans, Anabel De Proft, Barundeb Dutta, Carolina Mora Lopez imec, Heverlee, Belgium *Equally Credited Authors (ECAs) Abstract We present a waveguide-based optogenetics neural probe with 1,280 electrodes, 3
ISSCC 2026
Session 37
Memory
A 72Gb/s/pin Single-Ended Driver-Cooperative Coded PAM3 Transceiver with Asymmetric Data-Dependent Equalization and Bias-Peaking for Chiplets and Memory Interfaces
Abstract This paper presents a 72Gb/s/pin PAM-3 single-ended transceiver in 28nm CMOS. Drivercooperative coding enhances signal integrity and reduces signaling power by ≥46%. Asymmetric data-dependent equalization and bi
ISSCC 2026
Session 37
Memory
A 47.0Tb/s/mm 112Gb/s/pin PAM4 Single-Ended Transceiver Featuring 4-Aggressor Crosstalk Cancellation and Supply-Noise Tolerance for Short-Reach Memory Interfaces
Interdisciplinary Research Center for Future Intelligent Chips (Chip-X), Suzhou, China 1 3 Abstract This paper presents a five-lane 112Gb/s/pin PAM4 single-ended transceiver in 28nm CMOS for high-density short-reach memo
ISSCC 2026
Session 37
Memory
A 2nm All-Digital 14.4Gb/s/pin LPDDR6 PHY with Quarter-Rate Clocking Architecture and Multi-Level FIFO-Based Speculative DFE
LPDDR6 PHY that achieves 14.4Gb/s/pin, increasing bandwidth while optimizing power. Key innovations include quadrature clocking, with QEC, to reduce jitter sensitivity, and a multi-level FIFO-based speculative DFE to rel
ISSCC 2026
Session 37
Memory
A 16Gb/s/pin 0.51pJ/b Single-Ended NRZ Transceiver with Distributed Dual-Loop VDDQ-Ripple Compensation and Dynamic Clock Duty-Cycle Calibration for Memory Interfaces
*Equally Credited Authors (ECAs) 1 Abstract A 16Gb/s/pin 0.51pJ/b single-ended NRZ transceiver with distributed dual-loop compensation (DDLC) for VDDQ-ripple suppression and dynamic duty-cycle calibration (DDCC) for robu
ISSCC 2026
Session 37
Memory
A 0.092pJ/b and 7.7fJ/b/dB Cross-Self-Referenced Slope-Sampling Receiver with Long-Tail ISI Robustness for Next-Generation Low-Power Memory Interfaces
*Equally Credited Authors (ECAs) Abstract We present a cross-self-referenced slope-sampling RX that is robust against long-tail ISI, without equalization, for next-generation low-power memory interfaces. By determining r
ISSCC 2026
Session 37
Memory
A 12.8Gb/s Parallel Receiver with a One-Way Self-Training Scheme for Equalizing ISI and Reflections in Multi-Drop Memory Interfaces
8Gb/s/pin receiver with a one-way self-training scheme for highlyreflective multi-drop memory interfaces. DFE tap coefficients are directly derived from a single-pulse response, eliminating controller interaction during
ISSCC 2026
Session 37
Memory
A 0.87pJ/b 17Gb/s/pin Parallel Receiver with a Local DQS Recovery for a Supply-Noise-Tolerant DQS Distribution in High-Performance NAND Flash Interfaces
Abstract This work presents a 0.87pJ/b, 17Gb/s/pin parallel receiver with local DQS recovery and a dual-DQS tree for reduced clock power and mitigated power supply-induced jitter (PSIJ). The local DQS recovery time-multi
ISSCC 2026
Session 37
Memory
A 14Gb/s/pin 0.163pJ/b DQ Receiver for HBM with Baud-Rate Phase Tracking Loop Supporting Background Offset Calibration
Abstract This paper presents a compact 14Gb/s/pin HBM DQ receiver that continuously tracks sampling phase and detector offset in live data, using a baud-rate phase tracking loop with a background-calibrated time-window p
ISSCC 2026
Session 4
Analog Circuits
A 2.1-to-3.7ppm/°C Bandgap Voltage Reference with a Current-Domain TC Compensation and ±0.06% Inaccuracy from -40°C to 125°C in 130nm CMOS
Abstract This paper presents a CMOS bandgap reference (BGR) with a current-domain high-order TC compensation by using a capacitively-biased-diode-based super-PTAT current bias. It achieves a TC of 2.1 to 7.1ppm/°C from -
ISSCC 2026
Session 4
Analog Circuits
A 1ppm/°C and ±0.066% 3σ Accuracy Bandgap Reference with Temperature-Adaptive PTAT Scaling
Abstract This paper presents a process-independent, curvature-compensated bandgap reference. By introducing a temperature-adaptive duty-cycled resistor to scale the PTAT voltage, the design effectively compensates for CT
ISSCC 2026
Session 4
Analog Circuits
An Integrated Voltage and Current Reference Together Achieving 5.7 and 9.1ppm/°C from -40 to 125°C
Abstract This work implements an integrated voltage and current reference circuit, which has low temperature drift. By ensuring the reuse of the main circuit, we propose a current-mirrorbased seamless transition techniqu
ISSCC 2026
Session 5
mm-Wave
A Formation Flight Phased-Array Transceiver for Spatial Power Combining and Distributing Architectures in Direct-to-Device-Communication Satellite Constellations
Japan, 4Iwate University, Iwate, Japan 1 Abstract Conventional satellites are costly and unreliable. We propose a pico-satellite formation as a virtual phased array to reduce costs and improve reliability. A low-power FD
ISSCC 2026
Session 5
mm-Wave
A 28GHz Frequency-Diverse Sub-Array TX with Secret Phase Keys and Antenna Subset Modulation for Eavesdropping-Resilient Wireless Communication
Abstract Prior works on physical-layer eavesdropping protection only addressed off-axis Eve. This paper presents the first TX to defend against both main-beam and off-axis Eves. Incorporating three key components: freque
ISSCC 2026
Session 5
mm-Wave
A 140GHz Full-Duplex CMOS Transceiver with Metasurface-Integrated Self-Interference-Cancelling Antenna Supporting 16Gb/s 16-QAM Dual-Mode Bidirectional Communication
Abstract This paper presents a 140GHz CMOS transceiver for both full-duplex and frequency-division duplex communication. By proposing self-interference (SI)-cancelling antenna and dualpolarized metasurface, the design ac
ISSCC 2026
Session 6
mm-Wave
Full-Duplex RF Canceler Achieving Wideband High-SI-Power Low-Noise Cancellation Through A Novel N-Path-Filter-Based Architecture and ML-Based Canceler Configuration
Abstract We present a novel RF canceller that utilizes (i) a novel N-path-filter-based architecture that uses filter stacking to achieve passive tap combining leading to a lower NF degradation, (ii) a sub-cycle rotation
ISSCC 2026
Session 6
mm-Wave
A Phased-Array-Inspired Broadband RF Signal Processor for 2-to-32GHz Spectrum Sensing
Abstract This paper presents a phased-array-inspired RF signal processor for ultra-broadband spectrum sensing. The proposed architecture exploits the bandpass response resulting from the constructive and destructive inte
ISSCC 2026
Session 6
mm-Wave
A 2×2 10GS/s TTD BF Receiver Utilizing Charge-Based Summation with 10.5GHz Bandwidth and SNDR/SFDR with 49.5dB/57.5dBc in 22nm CMOS.
Abstract This work presents an ultra-compact 10GS/s sampled beamforming (BF) receiver front-end (FE) for a 2×2 antenna array, enabling both azimuth and elevation steering in 22nm FDSOI CMOS. The 4× time-interleaved FE fe
ISSCC 2026
Session 6
mm-Wave
A 436-to-472GHz 4-Element IF Beamforming Phased-Array Receiver in 65nm CMOS
Abstract This paper presents a 436-to-472GHz 4-element IF beamforming phased-array receiver in 65nm CMOS. A two-step mixing architecture reduces layout overhead. The LO frequency multiplier chain provides >-1dBm over 180
ISSCC 2026
Session 6
mm-Wave
A 26/28/37/39GHz Reconfigurable Fully Connected MIMO Receiver Front-End with On-Chip Diplexer Achieving 52-to-70dB Blocker Rejection
25-to-29.5 / 37-to-40GHz fully connected MIMO receiver (RX) front-end for 5G FR2 that enables 52-to-70dB calibration-free inter-band blocker rejection and up to -5dBm blocker tolerance is introduced. A compact on-chip di
ISSCC 2026
Session 7
Image Sensors
54×42 LiDAR 3D-Stacked System-On-Chip with On-Chip Point Cloud Processing and Hybrid On-Chip/Package-Embedded 25V Boost Generation
Duncan Hall1, Axel Crocherie1, Cedric Pastorelli1, Sophie Taupin2, Severin Trochut2, Abhishek Singh1, Andreas Assmann1, Bruce R. Rae1, Pascal Mellot2 STMicroelectronics, Edinburgh, United Kingdom, 2STMicroelectronics, Gr
ISSCC 2026
Session 7
Image Sensors
A 200MP 0.61μm-Pixel-Pitch CMOS Imager with Sub-1e- Readout Noise Using Interlaced-Shared Transistor Architecture and On-Chip Motion Artifact-Free HDR Synthesis for 8K Video Applications
Abstract We propose a 200MP 0.61µm pixel size CMOS image sensor featuring: 1) an interlacedshared pixel architecture to achieve a 0.7e- readout noise, 2) an on-chip motion artifact-free HDR synthesis with single-frame dy
ISSCC 2026
Session 7
Image Sensors
VoxCAD: A 0.82-to-81.0mW Intelligent 3D-Perception dToF SoC with Sector-Wise Voxelization and High-Density Tri-Mode eDRAM CIM Macro
*Equally Credited Authors (ECAs) 1 Abstract VoxCAD, an intelligent low-power dToF SoC for end-to-end 3D perception applications, is presented with three features: 1) LiDAR dToF sensing-integrated 2D-ROI guided point clou
ISSCC 2026
Session 7
Image Sensors
A 480×320 CMOS LiDAR Sensor with Tapering 1-Step Histogramming TDCs and Sub-Pixel Echo Resolvers
*Equally Credited Authors (ECAs) 1 Abstract A 480×320 CMOS LiDAR sensor using tapered 1-step hTDCs and SPER for memory-efficient operation is presented. The hTDC accumulates histograms with 1ns resolution over the full r
ISSCC 2026
Session 7
Image Sensors
A 26.0mW 30fps 400x300-pixel SWIR Ge-SPAD dToF Range Sensor with Programmable Macro-Pixels and Integrated Histogram Processing for Low-Power AR/VR Applications
Sony Semiconductor Solutions Europe, Trento, Italy, 2Sony Semiconductor Solutions, Atsugi, Japan 1 Abstract This paper presents a 400×300-pixel Ge-SPAD SWIR/NIR sensor enabling very-low-power operation: the flexible macr
ISSCC 2026
Session 7
Image Sensors
A 128×96 Multimodal Flash LiDAR SPAD Imager with Object Segmentation Latency of 18μs Based on Compute-Near-Sensor Ising Annealing Machine
*Equally-Credited Authors (ECAs) Abstract This paper presents a 128×96 multimodal flash LiDAR SPAD imager integrated with a compute-near-sensor Ising-model annealing processor for dynamic object segmentation, framed as a
ISSCC 2026
Session 7
Image Sensors
A Fully Reconfigurable Hybrid SPAD Vision Sensor with 134dB Dynamic Range Using Time-Coded Dual Exposures
Abstract This paper presents a fully reconfigurable HDR intensity/event hybrid SPAD sensor. The fully reconfigurable scheme enables both pixel types to be freely blended and operated simultaneously. A time-coded dual exp
ISSCC 2026
Session 7
Image Sensors
A 55nm Intelligent Vision SoC Achieving 346TOPS/W System Efficiency via Fully Analog Sensing-to-Inference Pipeline
presents a fully analog intelligent vision SoC that eliminates both sensorprocessor and inter-layer A/D conversions for end-to-end vision. A continuous analog datapath from sensing to multi-layer inference is enabled by
ISSCC 2026
Session 7
Image Sensors
A 1.09e--Random-Noise 1.5µm-Pixel-Pitch 12MP Global-Shutter-Equivalent CMOS Image Sensor with 3µm Digital Pixels Using Quad-Phase-Staggered Zigzag Readout and Motion Compensation
Gihwan Cho, Hyukbin Kwon, Jaehun Jeong, Bumjun Kim, Su-Hyun Han, Youna Lee, Hiroyuki Sugihara, Junghoon Jung, Suksan Kim, Kyungtae Lim, Wonoh Ryu, Yongjun Kim, Seung-Sik Kim, Heesung Shim, Min-Woong Seo, Jae-kyu Lee, Jon
ISSCC 2026
Session 8
Wireline I/O
A 48Gb/s/lane 1.24Tb/s/mm UCIe-Compliant Die-to-Die Link Over 30mm Standard Package
work demonstrates a UCIe-S compliant die-to-die PHY at 48Gb/s/lane across 16 lanes, achieving 1.24Tb/s/mm shoreline BW density over a 30mm organic package at 1.2pJ/b, extendable to 56GT/s (1.13pJ/b). Circuit innovations
ISSCC 2026
Session 8
Wireline I/O
A 180-to-240Gb/s Analog-Intensive PAM-4 Transmitter with 0.70pJ/b Analog Power Efficiency in 65nm CMOS
Abstract This paper presents a 180-240Gb/s analog-intensive PAM-4 transmitter in 65nm CMOS process. A three-stage cascaded 2-to-1 analog MUX (AMUX) is employed to reduce the complexity and therefore the parasitic capacit
ISSCC 2026
Session 8
Wireline I/O
A 1.59pJ/b 112Gb/s PAM-4 and 1.06pJ/b 168Gb/s PAM-8 Resistor-Less 7-Bit SST DAC-Based Transmitter with 8-Tap FFE in 28nm CMOS
Abstract A resistor-less 7b SST DAC-based TX with 8-tap FFE and 3 types of segments is presented to achieve low parasitic capacitance, compact area, and scaling friendliness. To reduce intersymbol-interference jitter at
ISSCC 2026
Session 8
Wireline I/O
A 32Gb/s 12.35Tb/s/mm2 0.36pJ/b UCIe-Like Die-to-Die Interface Featuring Edge-Triggered Transceivers in 3nm with Active LSI Packaging
Chin-Hua Wen1, Hsin-Hung Kuo1, Han-Tzung Ke1, Jie-Ren Huang1, Chang-Yi Li1, Sheng-Tsung Lai1, Shu-Chun Yang1, Kuan-Ting Chou1, Pei-Chen Chiou1, Tsung-Hsien Tsai1, Yi-Ting Chen1, Yen-Ming Chen1, Kenny Cheng-Hsiang Hsieh1
ISSCC 2026
Session 8
Wireline I/O
A 0.23pJ/b 24Gb/s Modular D2D Interface With Zero Wake Penalty Clock Gating in 3nm
Marina Salik2, Kevin Bartholomew2, Sushmitha Reddy2, Alex Tessitore2, Aws Shallal2, Kalyan Nallaparaju2, Jin Liang2, Minhan Chen2, Shaishav Desai1 Microsoft, Sunnyvale, CA, 2Microsoft, Raleigh, NC 1 Abstract This work pr
ISSCC 2026
Session 8
Wireline I/O
A 112Gb/s/wire Single-Ended Simultaneous Bi-Directional Transceiver with Dynamic Equalizer for Die-to-Die Interface in 28nm CMOS
*Equally Credited Authors (ECAs) 1 Abstract This work presents an 8-lane 112Gb/s/wire single-ended simultaneous bi-directional transceiver with a 3mm shield-less on-chip channel. A dynamic equalizer is proposed to decoup
ISSCC 2026
Session 8
Wireline I/O
A 112Gb/s 0.76pJ/b Reference-less Mixed-Signal PAM-4 CDR in 28nm CMOS
*Equally Credited Authors (ECAs) 1 Abstract This paper presents a 112Gb/s mixed-signal reference-less PAM-4 CDR in 28nm CMOS. By proposing the hybrid architecture based on a symmetrical linear PD and a bang-bang PFD, a f
ISSCC 2026
Session 8
Wireline I/O
A 280mW 112Gb/s PAM-4/NRZ Transceiver for Low-Power IOs in 5nm FinFET Technology
Alberto Grassi1, Mehrdad Fahimnia1, Hiroshi Kimura2, Faramarz Bahmani2, Hao Chen2, Alex Wang2, Chen Zhao2, Yuan Fang2, Allen Chen3, Afshin Momtaz1, Namik Kocaman1 Broadcom, Irvine, CA, 2Broadcom, San Jose, CA, 3Broadcom,
ISSCC 2026
Session 8
Wireline I/O
A 112Gb/s PAM-4 SBD Transceiver with Mismatch-Compensated 2×VDD Hybrid and Two-Step Echo Canceller in 28nm CMOS
Abstract A 112Gb/s PAM-4 simultaneous bidirectional (SBD) transceiver in 28nm CMOS is presented. It features a hybrid with a 2×VDD stacked driver to restore signal swing, a joint delay and slew-rate matching scheme to el
ISSCC 2026
Session 8
Wireline I/O
A 0.292pJ/b 56Gb/s/wire Capacitively Driven Simultaneous Bidirectional Transceiver with PVT/Mismatch Tracking for XSR and D2D Interfaces in 28nm CMOS
Abstract A low-power (0.292pJ/b), high-bandwidth (56Gb/s/wire) single-ended capacitively driven simultaneous bi-directional (CD-SBD) TRX with PVT tolerance is proposed. Capacitive driving reduces power and self-interfere
ISSCC 2026
Session 8
Wireline I/O
A 72Gb/s/pin Single-Ended Simultaneous Bi-Directional Transceiver with C-Peaking Leakage Cancellation and Dual-Loop Hybrid Impedance Calibration for Chiplet Interfaces
Abstract This paper presents a 72Gb/s/pin single-ended simultaneously bi-directional (SBD) TRX in 28nm CMOS. Capacitive peaking leakage cancellation (CPLC) suppresses the high-frequency leakage due to the main and hybrid
ISSCC 2026
Session 9
Wireless
A Single-Power-Link 13.56MHz Wireless Power and Data Transfer System with Synchronized Phase-Shifted Time-Multiplexing Dual Uplinks for Implantable Voltammetry
Abstract A 13.56MHz wireless power and data transfer (WPDT) system for implantable voltammetry sensor is presented. With the synchronized phase-shifted time-multiplexing technique, the system uses a single power link to
ISSCC 2026
Session 9
Wireless
A 91%-Efficiency Single-Stage Bipolar Quad-Output Regulating Rectifier with Event-Driven Output Power Enhancement via Coil-Reused DC-DC for Wireless Power Transfer
Abstract A bipolar quad-output regulating rectifier for wireless power transfer (WPT) is presented. Using only five power switches, it provides four independently regulated bipolar outputs. A coil-reused DC-DC mode susta
ISSCC 2026
Session 9
Wireless
An Output-Domain-Independent Single-Transmitter-Dual-Receiver Wireless Power Transfer System with Detuned-Tank and Time-Multiplexing Control for Adaptive Power Distribution
Abstract This paper presents an output-domain-independent single-transmitter-dual-receiver wireless power transfer system. The proposed detuned-tank and time-multiplexing control enable adaptive power distribution, preve
ISSCC 2026
Session 9
Wireless
A Multi-Coil Scalable Energy-Shared Wireless Power Receiver Network for Distributed Time-Division-Multiplexing Somatosensory Cortex Stimulation
Abstract This work presents a system-level scalable wireless power receiver (RX) network for miniaturized distributed somatosensory cortex stimulation. By interconnecting the power outputs of all RX cells in the network,
← 上一页 · 第 5/5 页 · 共 247 篇