Search for: All records

Artificial neural networks (ANNs) are widely used in numerous artificial intelligence‐based applications. However, the significant amount of data transferred between computing units and storage has limited the widespread deployment of ANN for the artificial intelligence of things (AIoT) and power‐constrained device applications. Therefore, among various ANN algorithms, quantized neural networks (QNNs) have garnered considerable attention because they require fewer computational resources with minimal energy consumption. Herein, an oxide‐based ternary charge‐trap transistor (CTT) that provides three discrete states and non‐volatile memory characteristics are introduced, which are desirable for QNN computing. By employing a differential pair of ternary CTTs, an artificial synaptic segregation with multilevel quantized values for QNNs is demostrated. The approach establishes a platform that combines the advantages of multiple states and robustness to noise for in‐memory computing to achieve reliable QNN performance in hardware, thereby facilitating the development of energy‐efficient AIoT.

 

As machine vision technology generates large amounts of data from sensors, it requires efficient computational systems for visual cognitive processing. Recently, in-sensor computing systems have emerged as a potential solution for reducing unnecessary data transfer and realizing fast and energy-efficient visual cognitive processing. However, they still lack the capability to process stored images directly within the sensor. Here, we demonstrate a heterogeneously integrated 1-photodiode and 1 memristor (1P-1R) crossbar for in-sensor visual cognitive processing, emulating a mammalian image encoding process to extract features from the input images. Unlike other neuromorphic vision processes, the trained weight values are applied as an input voltage to the image-saved crossbar array instead of storing the weight value in the memristors, realizing the in-sensor computing paradigm. We believe the heterogeneously integrated in-sensor computing platform provides an advanced architecture for real-time and data-intensive machine-vision applications via bio-stimulus domain reduction.

 

In‐sensor computing is an emerging architectural paradigm that fuses data acquisition and processing within a sensory domain. The integration of multiple functions into a single domain reduces the system footprint while it minimizes the energy and time for data transfer between sensory and computing units. However, it is challenging for a simple and compact image sensor array to achieve both sensing and computing in each pixel. Here, this work demonstrates a focal plane array with a heterogeneously integrated one‐photodiode one‐resistor (1P‐1R)‐based artificial optical neuron that emulates the sensing, computing, and memorization of a biological retina system. This work employs an InGaAs photodiode featuring a high responsivity and a broad spectrum that covers near‐infrared (NIR) signals and employs an HfO₂memristor as the artificial synapse to achieve the computing/memorization in an analog domain. Using the fabricated focal plane array integrated with an artificial neural network, this work performs in‐sensor image identification of finger veins driven by NIR light illumination (≈84 % accuracy). The proposed in‐sensor image computing architecture that broadly covers the NIR spectrum offers widespread application of focal plane array for computer vision, neuromorphic computing, biomedical engineering, etc.

 

A neuromuscular junction (NMJ) is a particularized synapse that activates muscle fibers for macro-motions, requiring more energy than computation. Emulating the NMJ is thus challenging owing to the need for both synaptic plasticity and high driving power to trigger motions. Here, we present an artificial NMJ using CuInP₂S₆(CIPS) as a gate dielectric integrated with an AlGaN/GaN-based high-electron mobility transistor (HEMT). The ferroelectricity of the CIPS is coupled with the two-dimensional electron gas channel in the HEMT, providing a wide programmable current range of 6 picoampere per millimeter to 5 milliampere per millimeter. The large output current window of the CIPS/GaN ferroelectric HEMT (FeHEMT) allows for amplifier-less actuation, emulating the biological NMJ functions of actuation and synaptic plasticity. We also demonstrate the emulation of biological oculomotor dynamics, including in situ object tracking and enhanced stimulus responses, using the fabricated artificial NMJ. We believe that the CIPS/GaN FeHEMT offers a promising pathway for bioinspired robotics and neuromorphic vision. 

Non-line-of-sight (NLOS) detection and ranging aim to identify hidden objects by sensing indirect light reflections. Although numerous computational methods have been proposed for NLOS detection and imaging, the post-signal processing required by peripheral circuits remains complex. One possible solution for simplifying NLOS detection and ranging involves the use of neuromorphic devices, such as memristors, which have intrinsic resistive-switching capabilities and can store spatiotemporal information. In this study, we employed the memristive spike-timing-dependent plasticity learning rule to program the time-of-flight (ToF) depth information directly into a memristor medium. By coupling the transmitted signal from the source with the photocurrent from the target object into a single memristor unit, we were able to induce a tunable programming pulse based on the time interval between the two signals that were superimposed. Here, this neuromorphic ToF principle is employed to detect and range NLOS objects without requiring complex peripheral circuitry to process raw signals. We experimentally demonstrated the effectiveness of the neuromorphic ToF principle by integrating a HfO2 memristor and an avalanche photodiode to detect NLOS objects in multiple directions. This technology has potential applications in various fields, such as automotive navigation, machine learning, and biomedical engineering. 

Significant effort for demonstrating a gallium nitride (GaN)-based ferroelectric metal–oxide–semiconductor (MOS)-high-electron-mobility transistor (HEMT) for reconfigurable operation via simple pulse operation has been hindered by the lack of suitable materials, gate structures, and intrinsic depolarization effects. In this study, we have demonstrated artificial synapses using a GaN-based MOS-HEMT integrated with an α-In2Se3 ferroelectric semiconductor. The van der Waals heterostructure of GaN/α-In2Se3 provides a potential to achieve high-frequency operation driven by a ferroelectrically coupled two-dimensional electron gas (2DEG). Moreover, the semiconducting α-In2Se3 features a steep subthreshold slope with a high ON/OFF ratio (∼1010). The self-aligned α-In2Se3 layer with the gate electrode suppresses the in-plane polarization while promoting the out-of-plane (OOP) polarization of α-In2Se3, resulting in a steep subthreshold slope (10 mV/dec) and creating a large hysteresis (2 V). Furthermore, based on the short-term plasticity (STP) characteristics of the fabricated ferroelectric HEMT, we demonstrated reservoir computing (RC) for image classification. We believe that the ferroelectric GaN/α-In2Se3 HEMT can provide a viable pathway toward ultrafast neuromorphic computing. 

Heterogeneous integration techniques allow the coupling of highly lattice-mismatched solid-state membranes, including semiconductors, oxides, and two-dimensional materials, to synergistically fuse the functionalities. The formation of heterostructures generally requires two processes: the combination of crystalline growth and a non-destructive lift-off/transfer process enables the formation of high-quality heterostructures. Although direct atomic interaction between the substrate and the target membrane ensures high-quality growth, the strong atomic bonds at the substrate/epitaxial film interface hinder the non-destructive separation of the target membrane from the substrate. Alternatively, a 2D material-coated compound semiconductor substrate can transfer the weakened (but still effective) surface potential field of the surface through the 2D material, allowing both high-quality epitaxial growth and non-destructive lift-off of the grown film. This Perspective reviews 2D/3D heterogeneous integration techniques, along with applications of III–V compound semiconductors and oxides. The advanced heterogeneous integration methods offer an effective method to produce various freestanding membranes for stackable heterostructures with unique functionalities that can be applied to novel electrical, optoelectronic, neuromorphic, and bioelectronic systems. 

3D sensing is a primitive function that allows imaging with depth information generally achieved via the time‐of‐flight (ToF) principle. However, time‐to‐digital converters (TDCs) in conventional ToF sensors are usually bulky, complex, and exhibit large delay and power loss. To overcome these issues, a resistive time‐of‐flight (R‐ToF) sensor that can measure the depth information in an analog domain by mimicking the biological process of spike‐timing‐dependent plasticity (STDP) is proposed herein. The R‐ToF sensors based on integrated avalanche photodiodes (APDs) with memristive intelligent matters achieve a scan depth of up to 55 cm (≈89% accuracy and 2.93 cm standard deviation) and low power consumption (0.5 nJ/step) without TDCs. The in‐depth computing is realized via R‐ToF 3D imaging and memristive classification. This R‐ToF system opens a new pathway for miniaturized and energy‐efficient neuromorphic vision engineering that can be harnessed in light‐detection and ranging (LiDAR), automotive vehicles, biomedical in vivo imaging, and augmented/virtual reality.

 

Defect identification has been a significant task in various fields to prevent the potential problems caused by imperfection. There is great attention for developing technology to accurately extract defect information from the image using a computing system without human error. However, image analysis using conventional computing technology based on Von Neumann structure is facing bottlenecks to efficiently process the huge volume of input data at low power and high speed. Herein efficient defect identification is demonstrated via a morphological image process with minimal power consumption using an oxide transistor and a memristor‐based crossbar array that can be applied to neuromorphic computing. Using a hardware and software codesigned neuromorphic system combined with a dynamic Gaussian blur kernel operation, an enhanced defect detection performance is successfully demonstrated with about 104 times more power‐efficient computation compared to the conventional complementary metal‐oxide semiconductor (CMOS)‐based digital implementation. It is believed the back end of line (BEOL)‐compatible all‐oxide‐based memristive crossbar array provides the unique potential toward universal artificial intelligence of things (AIoT) applications where conventional hardware can hardly be used.