Results 121 to 130 of about 353,149 (264)
By combining ionic nonvolatile memories and transistors, this work proposes a compact synaptic unit to enable low‐precision neural network training. The design supports in situ weight quantization without extra programming and achieves accuracy comparable to ideal methods. This work obtains energy consumption advantage of 25.51× (ECRAM) and 4.84× (RRAM)
Zhen Yang +9 more
wiley +1 more source
Oxide Semiconductor Thin‐Film Transistors for Low‐Power Electronics
This review explores the progress of oxide semiconductor thin‐film transistors in low‐power electronics. It illustrates the inherent material advantages of oxide semiconductor, which enable it to meet the low‐power requirements. It also discusses current strategies for reducing power consumption, including interface and structure engineering.
Shuhui Ren +8 more
wiley +1 more source
A compact neuromorphic synapse is presented, coupling anti‐ferroelectric capacitors with carbon nanotube devices to realize a non‐volatile, ternary STDP learning circuit. A calibrated compact model employs the negative differential resistance effect for ternary latching in a non‐volatile fashion.
Mohammad Khaleqi Qaleh Jooq +4 more
wiley +1 more source
A diagnostic method for reconfigurable intelligent surfaces (RIS) based on non‐uniform space‐time‐coding modulation is presented. Fault localization is achieved via amplitude‐only spectral measurements, eliminating the need for complex signal processing. A one‐to‐one mapping between harmonic components and RIS elements enables accurate detection.
Xiao Qing Chen +8 more
wiley +1 more source
Neural Information Processing and Time‐Series Prediction with Only Two Dynamical Memristors
The present study demonstrates how simple circuits with only two memristive devices are utilized to perform high complexity temporal information processing tasks, like neural spike detection in noisy environment, or time‐series prediction. This circuit simplicity is enabled by the dynamical complexity of the memristive devices, i.e.
Dániel Molnár +12 more
wiley +1 more source
A physics‐based compact model for Conductive‐Metal‐Oxide/HfOx ReRAM, accounting for ion dynamics, electronic conduction, and thermal effects, is presented. Accurate and versatile simulations of analog non‐volatile conductance modulation and memory state stabilization enable reliable circuit‐level studies, advancing the optimization of neuromorphic and ...
Matteo Galetta +9 more
wiley +1 more source
This review surveys oxide‐semiconductor devices for in‐memory and neuromorphic computing, highlighting recent progress and remaining challenges in charge‐trap, ferroelectric, and two‐transistor devices. Oxide semiconductors, featuring ultra‐low leakage, low‐temperature processing, and back‐end‐of‐line compatibility, are explored for analog in‐memory ...
Suwon Seong +4 more
wiley +1 more source
Concurrent Sensing and Communications Based on Intelligent Metasurfaces
An IM‐assisted concurrent ISAC scheme to exploit the otherwise wasted energy inherent in high‐order 16‐APSK DAM, to realize wireless gesture recognition thereby enabling efficient resource sharing for C‐ISAC. This architecture offers promising opportunities for next‐generation mobile internet, embodied intelligence, smart homes, smart factories, and ...
Hanting Zhao +8 more
wiley +1 more source
The Evolution of Gas Sensors Into Neuromorphic Systems
Gas sensors are vital for various applications, but conventional designs rely on separate sensing, memory, and processing units, limiting speed, power efficiency, and adaptability. Neuromorphic gas sensing overcomes these constraints by integrating all functions in a single device.
Kevin Dominguez +4 more
wiley +1 more source

