Search Results (415)

Phase-change memory (PCM) is a promising candidate for in-memory computation and neuromorphic computing due to its high endurance, low cycle-to-cycle variability, and low read noise. However, among other factors, its performance strongly depends on the post-lithography fabrication steps. This study examines the impact of reactive ion etching (RIE) on PCM device performance by evaluating different etching gas mixtures, CHF₃:O₂, H₂:Ar, and Ar, and determining their impact on key device characteristics, particularly initial resistance and cycling stability. The present study demonstrates that a two-step etching approach in which the capping layer is first removed using H₂:Ar and the underlying GST layer is subsequently etched using physical Ar sputtering ensures stable and reliable PCM operation. In contrast, chemically reactive gases negatively impact the initial resistance, cycling stability, and device lifetime, likely due to alterations in the material composition. For the cycling stability evaluation, an advanced measurement algorithm utilizing the aixMATRIX setup by aixACCT Systems is employed. This algorithm enables automated testing, dynamically adjusting biasing parameters based on cell responses, ensuring a stable ON/OFF ratio and high-throughput characterization. Full article

Traditional Hopfield neural networks (HNNs) suffer from limitations in generating controllable chaotic dynamics, which are essential for applications in neuromorphic computing and secure communications. Memristors, with their memory-dependent nonlinear characteristics, provide a promising approach to regulate neuronal activities, yet systematic studies on attractor offset behaviors remain scarce. In this study, we propose a fully memristive electromagnetic radiation neural network by incorporating three distinct memristors as external electromagnetic stimuli into an HNN. The parameters of the memristors were tuned to modulate chaotic oscillations, while variations in initial conditions were employed to explore multistability through bifurcation and basin stability analyses. The results demonstrate that the system enables large-scale amplitude control of chaotic signals via memristor parameter adjustments, allowing arbitrary scaling of attractor amplitudes. Various offset behaviors emerge, including parameter-driven symmetric double-scroll relocations in phase space and initial-condition-induced offset boosting that leads to extreme multistability. These dynamics were experimentally validated using an STM32-based electronic circuit, confirming precise amplitude and offset control. Furthermore, a multi-channel pseudo-random number generator (PRNG) was implemented, leveraging the initial-boosted offset to enhance security entropy. This offers a hardware-efficient chaotic solution for encrypted communication systems, demonstrating strong application potential. Full article

Event-based (neuromorphic) cameras depart from frame-based sensing by reporting asynchronous per-pixel brightness changes. This produces sparse, low-latency data streams with extreme temporal resolution but demands new processing paradigms. In this survey, we systematically examine neuromorphic vision along three main dimensions. First, we highlight the technological evolution and distinctive hardware features of neuromorphic cameras from their inception to recent models. Second, we review image-processing algorithms developed explicitly for event-based data, covering works on feature detection, tracking, optical flow, depth and pose estimation, and object recognition. These techniques, drawn from classical computer vision and modern data-driven approaches, illustrate the breadth of applications enabled by event-based cameras. Third, we present practical application case studies demonstrating how event cameras have been successfully used across various scenarios. Distinct from prior reviews, our survey provides a broader overview by uniquely integrating hardware developments, algorithmic progressions, and real-world applications into a structured, cohesive framework. This explicitly addresses the needs of researchers entering the field or those requiring a balanced synthesis of foundational and recent advancements, without overly specializing in niche areas. Finally, we analyze the challenges limiting widespread adoption, identify research gaps compared to standard imaging techniques, and outline promising directions for future developments. Full article

Spiking neural networks (SNNs) have emerged as a promising model for energy-efficient, event-driven processing of asynchronous event streams from Dynamic Vision Sensors (DVSs), a class of neuromorphic image sensors with microsecond-level latency and high dynamic range. Nevertheless, challenges persist in optimising training and effectively handling spatio-temporal complexity, which limits their potential for real-time applications on embedded sensing systems such as object tracking and recognition. Targeting this neuromorphic sensing pipeline, this paper proposes the Dynamic Tracking with Event Attention Spiking Network (DTEASN), a novel framework designed to address these challenges by employing a pure SNN architecture, bypassing conventional convolutional neural network (CNN) operations, and reducing GPU resource dependency, while tailoring the processing to DVS signal characteristics (asynchrony, sparsity, and polarity). The model incorporates two innovative, self-developed components: an event-driven multi-scale attention mechanism and a spatio-temporal event convolver, both of which significantly enhance spatio-temporal feature extraction from raw DVS events. An Event-Weighted Spiking Loss (EW-SLoss) is introduced to optimise the learning process by prioritising informative events and improving robustness to sensor noise. Additionally, a lightweight event tracking mechanism and a custom synaptic connection rule are proposed to further improve model efficiency for low-power, edge deployment. The efficacy of DTEASN is demonstrated through empirical results on event-based (DVS) object recognition and tracking benchmarks, where it outperforms conventional methods in accuracy, latency, event throughput (events/s) and spike rate (spikes/s), memory footprint, spike-efficiency (energy proxy), and overall computational efficiency under typical DVS settings. By virtue of its event-aligned, sparse computation, the framework is amenable to highly parallel neuromorphic hardware, supporting on- or near-sensor inference for embedded applications. Full article

Neuromorphic computing, inspired by the human brain’s architecture, offers a transformative approach to overcoming the limitations of traditional von Neumann systems by enabling highly parallel, energy-efficient information processing. Among emerging materials, MXenes—a class of two-dimensional transition metal carbides and nitrides—have garnered significant attention due to their exceptional electrical conductivity, tunable surface chemistry, and mechanical flexibility. This review comprehensively examines recent advancements in MXene-based optoelectronic synapses and neurons, focusing on their structural properties, device architectures, and operational mechanisms. We emphasize synergistic electrical–optical modulation in memristive and transistor-based synaptic devices, enabling improved energy efficiency, multilevel plasticity, and fast response times. In parallel, MXene-enabled optoelectronic neurons demonstrate integrate-and-fire dynamics and spatiotemporal information integration crucial for biologically inspired neural computations. Furthermore, this review explores innovative neuromorphic hardware platforms that leverage multifunctional MXene devices to achieve programmable synaptic–neuronal switching, enhancing computational flexibility and scalability. Despite these promising developments, challenges remain in device stability, reproducibility, and large-scale integration. Addressing these gaps through advanced synthesis, defect engineering, and architectural innovation will be pivotal for realizing practical, low-power optoelectronic neuromorphic systems. This review thus provides a critical roadmap for advancing MXene-based materials and devices toward next-generation intelligent computing and adaptive sensory applications. Full article

This study provides an in-depth analysis of the learning dynamics of multichannel optical neurons based on spatial solitons generated in lithium niobate crystals. Single-node and multi-node configurations with different topological complexities (3 × 3, 4 × 4, and 5 × 5) were compared, assessing how the number of channels, geometry, and optical parameters affect the speed and efficiency of learning. The simulations indicate that single-node neurons achieve the desired imbalance more rapidly and with lower energy expenditure, whereas multi-node structures require higher intensities and longer timescales, yet yield a greater variety of responses, more accurately reproducing the functional diversity of biological neural tissues. The results highlight how the plasticity of these devices can be entirely modulated through optical parameters, paving the way for fully optical photonic neuromorphic networks in which memory and computation are co-localized, with potential applications in on-chip learning, adaptive routing, and distributed decision-making. Full article

We report Al/ZnO/Al optoelectronic synaptic devices fabricated on c-plane and m-plane sapphire substrates using a sol–gel process. The devices exhibit essential synaptic behaviors such as excitatory postsynaptic current modulation, paired-pulse facilitation, and long-term learning–forgetting dynamics described by Wickelgren’s power law. Comparative analysis reveals that substrate orientation strongly influences memory performance: devices on m-plane consistently show higher EPSCs, slower decay rates, and superior retention compared to c-plane counterparts. These characteristics are attributed to crystallographic effects that enhance carrier trapping and persistent photoconductivity. To demonstrate their practical applicability, 3 × 3-pixel arrays of adjacent devices were constructed, where a “T”-shaped optical pattern was successfully encoded, learned, and retained across repeated stimulation cycles. These results highlight the critical role of substrate orientation in tailoring synaptic plasticity and memory retention, offering promising prospects for ZnO-based optoelectronic synaptic arrays in in-sensor neuromorphic computing and artificial visual memory systems. Full article

Recent advances in Radio Frequency (RF)-based device classification have shown promise in enabling secure and efficient wireless communications. However, the energy efficiency and low-latency processing capabilities of neuromorphic computing have yet to be fully leveraged in this domain. This paper is a first step toward enabling an end-to-end neuromorphic system for RF device classification, specifically supporting development of a neuromorphic classifier that enforces temporal causality without requiring non-neuromorphic classifier pre-training. This Spiking Neural Network (SNN) classifier streamlines the development of an end-to-end neuromorphic device classification system, further expanding the energy efficiency gains of neuromorphic processing to the realm of RF fingerprinting. Using experimentally collected WirelessHART transmissions, the TI-SNN achieves classification accuracy above 90% while reducing fingerprint density by nearly seven-fold and spike activity by over an order of magnitude compared to a baseline Rate-Encoded SNN (RE-SNN). These reductions translate to significant potential energy savings while maintaining competitive accuracy relative to Random Forest and CNN baselines. The results position the TI-SNN as a step toward a fully neuromorphic “RF Event Radio” capable of low-latency, energy-efficient device discrimination at the edge. Full article

Electrolyte-gated transistors (EGTs) have emerged as a highly promising platform for neuromorphic computing and bioelectronics, offering potential solutions to overcome the limitations of the von Neumann architecture. This comprehensive review examines recent advancements in EGT technology, focusing on three critical dimensions: materials, device configurations, and external field regulation strategies. We systematically analyze the development and properties of diverse electrolyte materials, including liquid electrolyte, polymer-based electrolytes, and inorganic solid-state electrolytes, highlighting their influence on ionic conductivity, stability, specific capacitance, and operational characteristics. The fundamental operating mechanisms of EGTs and electric double layer transistors (EDLTs) based on electrostatic modulation and ECTs based on electrochemical doping are elucidated, along with prevalent device configurations. Furthermore, the review explores innovative strategies for regulating EGT performance through external stimuli, including electric fields, optical fields, and strain fields/piezopotentials. These multi-field regulation capabilities position EGTs as ideal candidates for building neuromorphic perception systems and energy-efficient intelligent hardware. Finally, we discuss the current challenges such as material stability, interfacial degradation, switching speed limitations, and integration density. Furthermore, we outline future research directions, emphasizing the need for novel hybrid electrolytes, advanced fabrication techniques, and holistic system-level integration to realize the full potential of EGTs in next-generation computing and bio-interfaced applications. Full article

Two-dimensional (2D) materials are promising candidates for neuromorphic computing owing to their atomically thin structure and tunable optoelectronic properties. However, achieving controllable synaptic behavior via defect engineering remains challenging. In this work, we introduce X-ray irradiation as a facile strategy to modulate defect states and enhance synaptic plasticity in WSe₂-based optoelectronic synapses. The introduction of selenium vacancies via irradiation significantly improved both electrical and optical responses. Under electrical stimulation, short-term potentiation (STP) exhibited enhanced excitatory postsynaptic current (EPSC) retention exceeding 10%, measured 20 s after the stimulation peak. In addition, the nonlinearity of long-term potentiation (LTP) and long-term depression (LTD) was reduced, and the signal decay time was extended. Under optical stimulation, STP showed more than 4% improvement in EPSC retention at 16 s with similar relaxation enhancement. These effects are attributed to irradiation-induced defect states that facilitate charge carrier trapping and extend signal persistence. Moreover, the reduced nonlinearity in synaptic weight modulation improved the recognition accuracy of handwritten digits in a CrossSim-simulated MNIST task, increasing from 88.5% to 93.75%. This study demonstrates that X-ray irradiation is an effective method for modulating synaptic weights in 2D materials, offering a universal strategy for defect engineering in neuromorphic device applications. Full article

The rapid growth of artificial intelligence (AI), data centers, and high-performance computing (HPC) has increased the demand for large bandwidth, high energy efficiency, and high-density optical interconnects. Co-packaged optics (CPO) technology offers a promising solution by integrating photonic integrated circuits (PICs) directly within or close to electronic integrated circuit (EIC) packages. This paper explores the evolution of CPO performance from various perspectives, including fan-out wafer level packaging (FOWLP), through-silicon via (TSV)-based packaging, through-glass via (TGV)-based packaging, femtosecond laser direct writing waveguides, ion-exchange glass waveguides, and optical coupling. Micro ring resonators (MRRs) are a high-density integration solution due to their compact size, excellent energy efficiency, and compatibility with CMOS processes. However, traditional thermal tuning methods face limitations such as high static power consumption and severe thermal crosstalk. To address these issues, non-volatile neuromorphic photonics has made breakthroughs using phase-change materials (PCMs). By combining the integrated storage and computing capabilities of photonic memory with the efficient optoelectronic interconnects of CPO, this deep integration is expected to work synergistically to overcome material, integration, and architectural challenges, driving the development of a new generation of computing hardware with high energy efficiency, low latency, and large bandwidth. Full article

For most, if not all, AI-accelerated hardware, communication with the agent is expensive and heavily bottlenecks the hardware performance. This omnipresent hardware restriction is also found in neuromorphic computing: a novel style of computing that involves deploying spiking neural networks to specialized hardware to achieve low size, weight, and power (SWaP) compute. In neuromorphic computing, spike trains, times, and values are used to communicate information to, from, and within the spiking neural network. Input data, in order to be presented to a spiking neural network, must first be encoded as spikes. After processing the data, spikes are communicated by the network that represent some classification or decision that must be processed by decoder logic. In this paper, we first present principles for interconverting between spike trains, times, and values using custom-designed spiking subnetworks. Specifically, we present seven networks that encompass the 15 conversion scenarios between these encodings. We then perform three case studies where we either custom design a novel network or augment existing neural networks with these conversion subnetworks to vastly improve their communication performance with the outside world. We employ a classic space vs. time tradeoff by pushing spike data encoding and decoding techniques into the network mesh (increasing space) in order to minimize intra- and extranetwork communication time. This results in a classification inference speedup of 23× and a control inference speedup of 4.3× on field-programmable gate array hardware. Full article

In the era of artificial intelligence, neuromorphic devices that simulate brain functions have received increasingly widespread attention. In this paper, an artificial neural synapse device based on ZnAlSnO thin-film transistors was fabricated, and its electrical properties were tested: the current-switching ratio was 1.18 × 10⁷, the subthreshold oscillation was 1.48 V/decade, the mobility was 2.51 cm²V⁻¹s⁻¹, and the threshold voltage was −9.40 V. Stimulating artificial synaptic devices with optical signals has the advantages of fast response speed and good anti-interference ability. The basic biological synaptic characteristics of the devices were tested under 365 nm light stimulation, including excitatory postsynaptic current (EPSC), paired-pulse facilitation (PPF), short-term plasticity (STP), and long-term plasticity (LTP). This device shows good synaptic plasticity. In addition, by changing the gate voltage, the excitatory postsynaptic current of the device at different gate voltages was tested, two different logical operations of “AND” and “OR” were achieved, and the influence of different synaptic states on memory was simulated. This work verifies the application potential of the device in the integrated memory and computing architecture, which is of great significance for promoting the high-quality development of neuromorphic computing hardware. Full article

Gallium oxide (Ga₂O₃)-based memristors are gaining traction as promising candidates for next-generation electronic devices toward in-memory computing, leveraging the unique properties of Ga₂O₃, such as its wide bandgap, high thermodynamic stability, and chemical stability. This review explores the evolution of memristor theory for Ga₂O₃-based materials, emphasising capacitive memristors and their ability to integrate resistive and capacitive switching mechanisms for multifunctional performance. We discussed the state-of-the-art fabrication methods, material engineering strategies, and the current challenges of Ga₂O₃-based memristors. The review also highlights the applications of these memristors in memory technologies, neuromorphic computing, and sensors, showcasing their potential to revolutionise emerging electronics. Special focus has been placed on the use of Ga₂O₃ in capacitive memristors, where their properties enable improved switching speed, endurance, and stability. In this paper we provide a comprehensive overview of the advancements in Ga₂O₃-based memristors and outline pathways for future research in this rapidly evolving field. Full article

Event-based cameras, as a revolutionary class of dynamic vision sensors, offer transformative advantages for capturing transient mechanical phenomena through their asynchronous, per-pixel brightness change detection mechanism. These neuromorphic sensors excel in challenging industrial environments with their microsecond-level temporal resolution, ultra-low power requirements, and exceptional dynamic range that significantly outperform conventional imaging systems. In this way, the event-based camera provides a promising tool for machine vibration sensing and fault diagnosis. However, the dynamic vision data from the event-based cameras have a complex structure, which cannot be directly processed by the mainstream methods. This paper proposes a dynamic vision-based non-contact machine fault diagnosis method. The Eagle Vision Transformer (EViT) architecture is proposed, which incorporates biologically plausible computational mechanisms through its innovative Bi-Fovea Self-Attention and Bi-Fovea Feedforward Network designs. The proposed method introduces an original computational framework that effectively processes asynchronous event streams while preserving their inherent temporal precision and dynamic response characteristics. The proposed methodology demonstrates exceptional fault diagnosis performance across diverse operational scenarios through its unique combination of multi-scale spatiotemporal feature analysis, adaptive learning capabilities, and transparent decision pathways. The effectiveness of the proposed method is extensively validated by the practical condition monitoring data of rotating machines. By successfully bridging cutting-edge bio-inspired vision processing with practical industrial monitoring requirements, this work creates a new paradigm for dynamic vision-based non-contact machinery fault diagnosis that addresses critical limitations of conventional approaches. The proposed method provides new insights for predictive maintenance applications in smart manufacturing environments. Full article