Transition-Metal Nitride Halide Dielectrics for Transition-Metal Dichalcogenide Transistors


Abstract "Using first-principles calculations, we investigate six transition-metal nitride halides (TMNHs): HfNBr, HfNCl, TiNBr, TiNCl, ZrNBr, and ZrNCl as potential van der Waals (vdW) dielectrics for transition metal dichalcogenide (TMD) channel transistors. We calculate the exfoliation energies and bulk phonon energies and find that the six TMNHs are exfoliable and thermodynamically stabl... » read more

An Event-Driven and Fully Synthesizable Architecture for Spiking Neural Networks


Abstract:  "The development of brain-inspired neuromorphic computing architectures as a paradigm for Artificial Intelligence (AI) at the edge is a candidate solution that can meet strict energy and cost reduction constraints in the Internet of Things (IoT) application areas. Toward this goal, we present μBrain: the first digital yet fully event-driven without clock architecture, with co-lo... » read more

NeuroSim Simulator for Compute-in-Memory Hardware Accelerator: Validation and Benchmark


Abstract:   "Compute-in-memory (CIM) is an attractive solution to process the extensive workloads of multiply-and-accumulate (MAC) operations in deep neural network (DNN) hardware accelerators. A simulator with options of various mainstream and emerging memory technologies, architectures, and networks can be a great convenience for fast early-stage design space exploration of CIM hardw... » read more

Toward Software-Equivalent Accuracy on Transformer-Based Deep Neural Networks With Analog Memory Devices


Abstract:  "Recent advances in deep learning have been driven by ever-increasing model sizes, with networks growing to millions or even billions of parameters. Such enormous models call for fast and energy-efficient hardware accelerators. We study the potential of Analog AI accelerators based on Non-Volatile Memory, in particular Phase Change Memory (PCM), for software-equivalent accurate i... » read more

Benchmarking Highly Parallel Hardware for Spiking Neural Networks in Robotics


Abstract: "Animal brains still outperform even the most performant machines with significantly lower speed. Nonetheless, impressive progress has been made in robotics in the areas of vision, motion- and path planning in the last decades. Brain-inspired Spiking Neural Networks (SNN) and the parallel hardware necessary to exploit their full potential have promising features for robotic applica... » read more

Next Steps For Panel-Level Packaging


Tanja Braun, group manager at Fraunhofer Institute for Reliability and Microintegration (IZM), sat down with Semiconductor Engineering to talk about III-V device packaging, chiplets, fan-out and panel-level processing. Fraunhofer IZM recently announced a new phase of its panel-level packaging consortium. What follows are excerpts of that discussion. SE: IC packaging isn’t new, but years a... » read more

Smart Glasses, Augmented Reality, And Time-of-Flight Imaging


By Edmund Neo and Rolf Weber Augmented reality (AR) is frequently discussed for providing new, exciting, vision-related products for consumer and industrial applications. One popular AR implementation uses AR glasses. While the global market for smart augmented reality glasses was only 255,600 units in 2020, market researchers predict robust growth and a volume of 8.8 million units by 2026 [... » read more

Perspectives On Why EUV Photomasks Are More Expensive


There are fewer photomasks per wafer using EUV lithography, but each EUV photomask is more expensive. Given that, it’s not a surprise that a majority (74%) of industry luminaries surveyed in July say that EUV photomasks will contribute to an increase in photomask revenues for 2021 as shown in figure 1. In a 20-minute video, a panel of experts share their perspectives on what drives EUV photom... » read more

Week In Review: Manufacturing, Test


Chipmakers Chip investments in Malaysia got a shot in the arm this week. First, Intel has announced plans to invest more than RM30 billion, or US$7 billion, within its Malaysian packaging and test facilities. The additional investment will help expand Intel Malaysia’s operations across Penang and Kulim. This new investment is expected to create over 4,000 Intel jobs as well as over 5,000 con... » read more

Week In Review: Design, Low Power


Memory CEA-Leti demonstrated 16-kbit ferroelectric random-access memory (FeRAM) arrays at the 130nm node. It utilizes back-end-of-line (BEOL) integration of TiN/HfO2:Si/TiN ferroelectric capacitors as small as 0.16 µm² and solder reflow compatibility for the first time for this type of memory. The researchers anticipate it will be useful for embedded applications such at IoT and wearable dev... » read more

← Older posts Newer posts →