Using High-Quality Deterministic Patterns For In-System/In-Field Testing


Logic BIST (LBIST) is a well-stablished traditional solution for meeting automotive testing standards. However, using pseudo-random LBIST patterns can be challenging when trying to achieve high-quality testing due to the increased complexity of designs. The growing amount of electronic content, along with the shift toward fully autonomous vehicles, demands stringent testing requirements. In-... » read more

Detecting Slips, Scratches, Cracks In Wafers And Dies Becoming Harder


Defect detection requirements on the order of 10 defective parts per million (DPPM) are driving improvements in inspection tools’ resolution and throughput at foundries and OSATs. However, defects that manifest as slips, scratches, and micro-cracks continue to bedevil the prevalent optical inspection methods. These defects can range in size from nanometers to millimeters, some of which are... » read more

GaN-on-Si Technology Powers The Future Of Wireless


Mobile data traffic continues to grow. Constant video streaming is on the rise in applications like TikTok, Netflix, and Instagram, while emerging AI-enhanced mobile apps are set to further boost data consumption by offering highly personalized experiences and advanced features like augmented reality and voice processing in real-time. Ericsson estimates that mobile data traffic, excluding fixed... » read more

Solving The Mixed Criticality Challenge In Automotive Controllers


Car users want immersive and interactive in-vehicle experiences, especially as autonomous driving technology takes over more of our driving responsibilities. To satisfy this demand, automotive manufacturers are deploying multiple display environments in their cars, from cockpits to heads-up displays and from central infotainment screens to rear-seat entertainment. All of these displays are powe... » read more

Cache-Coherent Symmetric Multiprocessing With LX8 Controllers On HiFi DSPs


Overview This document motivates the need for a cache-coherent multicore, symmetric multiprocessor (SMP) with applications to embedded control and audio processing. It describes the current state of the art in the employment of multiple embedded processors and audio DSPs in an SoC, with a discussion of related problems that current designs face. This is followed by an introdu... » read more

Enhancing Dexterous Humanoid Robot Hands With Superior Touch ICs


New Paths to Human-Level Dexterity Humanoid robots and other general-purpose robotic platforms have made astonishing progress in locomotion and perception. Some can now run, jump, and even perform flips, while advanced computer vision enables object recognition at scale. Yet true human-level dexterity requires more than just bipedal agility or high-resolution cameras: it demands a refined s... » read more

PUFs In A Post-Quantum World


With the looming threat of quantum computing on the horizon, the security landscape is changing. Explore the emerging threat and its implications for current cryptographic standards. This white paper provides an in-depth analysis of quantum computing's impact on security and explains how PUF technology can help you maintain robust security in the quantum era. Why Read This? Quantum Comp... » read more

EDA Startups At DAC 2025


The 62nd DAC showcased numerous new exhibitors in 2025, including tool and IP providers, design services firms, and component marketplaces. New EDA startups, in particular, had a robust showing, with entrepreneurial engineers seeking to tackle the increasingly complex challenges facing modern chip design with fresh approaches. AI was a strong theme throughout the show, with companies of all ... » read more

10Base-T1S Ethernet And The Use Of MACsec For Link Security


10Base-T1S (henceforth in this blog referred to as T1S) is a relatively new standard for a 10Mbit/s single-pair Ethernet over an unshielded, single-twisted pair that was introduced with the IEEE 802.3cg-2019 specification. It brings Ethernet technology into the low-cost, edge applications space, enabling connection of “edge nodes” (remote sensors, actuators, light controls, etc.) to a “he... » read more

Shrinking LLMs With Self-Compression


Language models are becoming ever larger, making on-device inference slow and energy-intensive. A direct and surprisingly effective remedy is to prune complete channels whose contribution to the task is negligible. Our earlier work introduced a training-time procedure – Self-Compression [1, 4] – that lets back-propagation decide the bit-width of every channel, so unhelpful ones fade away. T... » read more

← Older posts Newer posts →