Category: Hardware

Researchers pioneer next-generation AI semiconductors with ‘thermal constraining’ technique A research team led by Professor Taesung Kim from the School of Mechanical Engineering at Sungkyunkwan University has developed a technology that precisely controls the internal structure of semiconductors using heat, much like stamping out “bungeoppang” (fish-shaped pastry) in a mold. The team report that this…

Miniaturized radar chip developed for next-generation wireless networks A miniaturized radar chip, developed by researchers at Science Tokyo, advances Integrated Sensing and Communication for Beyond 5G and 6G systems. Measuring just 0.24 mm2 and consuming only 9.8 mW, the compact, low-power device generates high-speed, highly linear chirps by embedding linearization directly into the hardware, overcoming…

Atom-thin ferroelectric transistor can store 3,024 polarization states Over the past few decades, electronics engineers have been trying to develop new neuromorphic hardware, systems that mirror the organization of neurons in the human brain. These systems could run artificial intelligence (AI) models, particularly artificial neural networks (ANNs) more reliably and efficiently than existing devices. Go…

New smart chip reduces consumption and computing time, advancing high-performance computing A new chip aims to dramatically reduce energy consumption while accelerating the processing of large amounts of data. Go to techxplore

New memristor training method slashes AI energy use by six orders of magnitude In a Nature Communications study, researchers from China have developed an error-aware probabilistic update (EaPU) method that aligns memristor hardware’s noisy updates with neural network training, slashing energy use by nearly six orders of magnitude versus GPUs while boosting accuracy on vision…

Reinforcement learning accelerates model-free training of optical AI systems Optical computing has emerged as a powerful approach for high-speed and energy-efficient information processing. Diffractive optical networks, in particular, enable large-scale parallel computation through the use of passive structured phase masks and the propagation of light. However, one major challenge remains: systems trained in model-based simulations…

Turning PCs and mobile devices into AI infrastructure can slash operational costs Until now, AI services based on large language models (LLMs) have mostly relied on expensive data center GPUs. This has resulted in high operational costs and created a significant barrier to entry for utilizing AI technology. A research team at KAIST has developed…

AI headphones automatically learn who you’re talking to—and let you hear them better Holding a conversation in a crowded room often leads to the frustrating “cocktail party problem,” or the challenge of separating the voices of conversation partners from a hubbub. It’s a mentally taxing situation that can be exacerbated by hearing impairment. Go to…

Engineers develop thin film to make AI chips faster and more energy efficient Addressing the staggering power and energy demands of artificial intelligence, engineers at the University of Houston have developed a revolutionary new thin-film material that promises to make AI devices significantly faster while dramatically cutting energy consumption. Go to techxplore

Integrated photonic millimeter-wave radar chip developed for next-generation radar 6G networks In a breakthrough in radar technology, researchers from the City University of Hong Kong (CityUHK) have developed the world’s first integrated photonic millimeter-wave radar chip, achieving unprecedented precision in a remarkably compact device. This breakthrough represents a significant step forward in the development of…

Beyond electronics: Optical system performs feature extraction with unprecedented low latency Many modern artificial intelligence (AI) applications, such as surgical robotics and real-time financial trading, depend on the ability to quickly extract key features from streams of raw data. This process is currently bottlenecked by traditional digital processors. The physical limits of conventional electronics prevent…

AI tools can help hackers plant hidden flaws in computer chips, study finds Widely available artificial intelligence systems can be used to deliberately insert hard-to-detect security vulnerabilities into the code that defines computer chips, according to new research from the NYU Tandon School of Engineering, a warning about the potential weaponization of AI in hardware…

Semiconductor neuron mimics brain’s memory and adaptive response abilities The human brain does more than simply regulate synapses that exchange signals; individual neurons also process information through intrinsic plasticity, the adaptive ability to become more sensitive or less sensitive depending on context. Existing artificial intelligence semiconductors, however, have struggled to mimic this flexibility of the…

Material that listens: Chip-based approach enables speech recognition and more Speech recognition without heavy software or energy-hungry processors: researchers at the University of Twente, together with IBM Research Europe and Toyota Motor Europe, present a completely new approach. Their chips allow the material itself to “listen.” The publication by Prof. Wilfred van der Wiel and…

Silent deep bass: Wearable audio you can feel Researchers at University of Tsukuba have developed a portable, silent subwoofer that combines electrical muscle stimulation with low-frequency vibrations. This device enables users to physically feel deep bass in virtual reality (VR) and everyday music. While minimizing noise, it provides an immersive experience and rhythm perception comparable…

New non-volatile memory platform built with covalent organic frameworks Researchers at Institute of Science Tokyo have created a new material platform for non-volatile memories using covalent organic frameworks (COFs), which are crystalline solids with high thermal stability. The researchers successfully installed electric-field-responsive dipolar rotors into COFs. Go to techxplore

Scientist discusses development of artificial synapses that mimic human brain function for next-gen AI chips The Emerging Investigator Series by the journal Materials Horizons features outstanding work by young researchers in the field of materials science. In the latest Editorial, Dr. Eunho Lee, an Assistant Professor of Chemical and Biomolecular Engineering at Seoul National University…

Oxygen defects help unlock the secret of next-generation memory Resistive random access memory (ReRAM), which is based on oxide materials, is gaining attention as a next-generation memory and neuromorphic computing device. Its fast speeds, data retention ability, and simple structure make it a promising candidate to replace existing memory technologies. Go to techxplore

Scientists develop the world’s first 6G chip, capable of 100 Gbps speeds Sixth generation, or 6G, wireless technology is one step closer to reality with news that Chinese researchers have unveiled the world’s first “all-frequency” 6G chip. The chip is capable of delivering mobile internet speeds exceeding 100 gigabits per second (Gbps) and was developed…

Artificial neuron merges DRAM with MoS₂ circuits to better emulate brain-like adaptability The rapid advancement of artificial intelligence (AI) and machine learning systems has increased the demand for new hardware components that could speed up data analysis while consuming less power. As machine learning algorithms draw inspiration from biological neural networks, some engineers have been…

AI cloud infrastructure gets faster and greener: NPU core improves inference performance by over 60% The latest generative AI models such as OpenAI’s ChatGPT-4 and Google’s Gemini 2.5 require not only high memory bandwidth but also large memory capacity. This is why generative AI cloud operating companies like Microsoft and Google purchase hundreds of thousands…

New framework reduces memory usage and boosts energy efficiency for large-scale AI graph analysis BingoCGN, a scalable and efficient graph neural network accelerator that enables inference of real-time, large-scale graphs through graph partitioning, has been developed by researchers at the Institute of Science Tokyo, Japan. This breakthrough framework utilizes an innovative cross-partition message quantization technique…

3D chip stacking method created to overcome traditional semiconductor limitations A novel power supply technology for 3D-integrated chips has been developed by employing a three-dimensionally stacked computing architecture consisting of processing units placed directly above stacks of dynamic random access memory. Go to techxplore

New all-silicon computer vision hardware advances in-sensor visual processing technology Researchers at the University of Massachusetts Amherst have pushed forward the development of computer vision with new, silicon-based hardware that can both capture and process visual data in the analog domain. Their work, described in the journal Nature Communications, could ultimately add to large-scale, data-intensive…

Tiny receiver chip uses stacked capacitors to block interference in 5G IoT devices MIT researchers have designed a compact, low-power receiver for 5G-compatible smart devices that is about 30 times more resilient to a certain type of interference than some traditional wireless receivers. Go to techxplore

Photonic processor could streamline 6G wireless signal processing As more connected devices demand an increasing amount of bandwidth for tasks like teleworking and cloud computing, it will become extremely challenging to manage the finite amount of wireless spectrum available for all users to share. Go to techxplore

Chinese researchers debut world’s first AI-based processor chip design system A team of engineers, AI specialists and chip design researchers at the Chinese Academy of Sciences has designed, built and tested what they are describing as the first AI-based chip design system. The group has published a paper describing their system, called QiMeng, on the…

‘Optical neural engine’ can solve partial differential equations Partial differential equations (PDEs) are a class of mathematical problems that represent the interplay of multiple variables, and therefore have predictive power when it comes to complex physical systems. Solving these equations is a perpetual challenge, however, and current computational techniques for doing so are time-consuming and…

AI chip developed for decentralized use without the cloud A new AI chip developed at the Technical University of Munich (TUM) works without the cloud server or internet connections needed by existing chips. The AI Pro, designed by Prof Hussam Amrouch, is modeled on the human brain. Its innovative neuromorphic architecture enables it to perform…

More efficient method developed to monitor drivers’ attention levels Fifty-two people have died on Catalan roads so far this year, according to data from the Catalan Traffic Service. Although there have been fewer accidents in the first four months of the year than in the same period last year, accidents continue to have a terrible…

Dual scalable annealing processors overcome capacity and precision limits Combinatorial optimization problems (COPs) arise in various fields such as shift scheduling, traffic routing, and drug development. However, they are challenging to solve using traditional computers in a practical timeframe. Go to techxplore

A shortcut to AI computation: In-memory computing overcomes data transfer bottlenecks As artificial intelligence (AI) continues to advance, researchers at POSTECH (Pohang University of Science and Technology) have identified a breakthrough that could make AI technologies faster and more efficient. Go to techxplore

Novel out-of-core mechanism introduced for large-scale graph neural network training A research team has introduced a new out-of-core mechanism, Capsule, for large-scale GNN training, which can achieve up to a 12.02× improvement in runtime efficiency, while using only 22.24% of the main memory, compared to SOTA out-of-core GNN systems. This work was published in the…

Researchers develop smaller, more efficient space-tolerant computer chips Space is a highly volatile environment. Factors like radiation, extreme temperatures, and debris make outer space a challenging environment for operating technology. In particular, radiation can have devastating effects on computer chips. Go to techxplore

Programmable photonic chip uses light to accelerate AI training and cut energy use Penn Engineers have developed the first programmable chip that can train nonlinear neural networks using light—a breakthrough that could dramatically speed up AI training, reduce energy use and even pave the way for fully light-powered computers. Go to techxplore

Photonic chips boost computing speed and efficiency to address growing demand Computer chips that combine the use of light and electricity are shown to increase computational performance, while reducing energy consumption, compared with conventional electronic chips. The photonic computing chips, described in two papers in Nature this week, might address the growing computing demands driven…

Korean researchers develop a processor for real-time hologram generation Korean researchers have developed a digital holography processor that converts two-dimensional (2D) videos into real-time three-dimensional (3D) holograms. This technology is expected to play a key role in the future of holography, as it enables the instantaneous transformation of ordinary 2D videos into 3D holograms. Go…

To keep hardware safe, new ‘Oreo’ method cuts out the code’s clues Imagine you’re a chef with a highly sought-after recipe. You write your top-secret instructions in a journal to ensure you remember them, but its location within the book is evident from the folds and tears on the edges of that often-referenced page. Go…

Sustainable SOT-MRAM memory technology could replace cache memory in computer architecture in the future How much energy is consumed each time we upload an image to social media, which relies on data centers and cloud storage? Data centers currently account for about 1% of global energy consumption, amounting to 200 terawatt-hours of electricity annually. This…

Big Tech wants to plug data centers right into power plants. Utilities say it’s not fair Looking for a quick fix for their fast-growing electricity diets, tech giants are increasingly looking to strike deals with power plant owners to plug in directly, avoiding a potentially longer and more expensive process of hooking into a fraying…

Ultra-small neuromorphic chip learns and corrects errors autonomously Existing computer systems have separate data processing and storage devices, making them inefficient for processing complex data like AI. A KAIST research team has developed a memristor-based integrated system similar to the way our brain processes information. It is now ready for application in various devices, including…

Specialized hardware solves high-order optimization problems with in-memory computing The rise of AI, graphic processing, combinatorial optimization and other data-intensive applications has resulted in data-processing bottlenecks, as ever greater amounts of data must be shuttled back and forth between the memory and compute elements in a computer. The physical distance is small, but the process…

Researchers from the Fraunhofer Institute for Photonic Microsystems IPMS, in collaboration with HOLOEYE Photonics AG, have developed a compact LCOS microdisplay with high refresh rates that enables improved optical modulation. This innovative microdisplay will be presented for the first time at the 31st International Display Workshops (IDW 2024) held Dec. 4–6 in Sapporo, Japan.