Blog Review: Nov. 11

UVM transaction class methods; EPROM; WPA2 vulnerabilities; NN inference on mobile GPUs.

popularity

Mentor’s Chris Spear proposes mixing together the compactness of the field macro style with the preciseness of the do methods when writing a UVM transaction class.

Cadence’s Paul McLellan looks back at the history of EPROM, some of the difficulty with actually erasing it, and the subsequent development of EEPROM.

Synopsys’ Tuomo Untinen explains three WPA2 authentication vulnerabilities recently found in wireless routers and investigates whether they are present in new WPA3 devices.

Ansys’ Wim Slagter provides some tips on selecting the best processor, memory, storage, and networking to match your simulation needs.

Arm’s Roberto Lopez Mendez explores the company’s open-source neural network inference engine, Arm NN, and how it uses OpenCL to improve inference on mobile GPUs.

In a video, VLSI Research’s Dan Hutcheson chats with Steve Pateras of Synopsys about the task of managing a chip from fab to the system to the trash bin using on-chip monitors and sensors to optimize all phases of the silicon lifecycle.

In a blog for SEMI, Jean-Marc Philippe of STMicroelectronics Singapore shares the experience of improving aging wafer equipment by adding Industry 4.0 technologies like real-time fault detection and classification combined with in-house designed sensors.

Memory blogger Jim Handy explains why 3D NAND is stuck at 40nm and the process steps that make further shrinkage nearly impossible.

A Rambus writer checks out the Open Compute Project’s Root of Trust (RoT) specification version 1.0, released as part of its work tackling the challenges of data security in the cloud.

Intel’s Edward Dixon ponders some ethical issues related to autonomous driving, artificial intelligence, and materials sourcing for manufacturing.

Silicon Labs’ Jackie Padgett points to eight principles that can be used to improve IoT security, from eliminating universal passwords and unsecured interfaces to clearly communicating when security updates will end.

NXP’s Sara Ellinger considers whether fingerprints will be the new authentication method for payment and government ID cards, instead of typical signatures or PINs.

And don’t miss the blogs from the latest Automotive, Security, & Pervasive Computing and Test, Measurement & Analytics newsletters:

Editor in Chief Ed Sperling observes that carmakers are beginning to think more realistically about electronics.

Rambus’ Paul Karazuba demonstrates improving device security with unique identification, secure boot, and protection of sensitive debug functions.

Flex Logix’s Geoff Tate explains why the size of images being processed changes how accelerators should be assessed.

Synopsys’ Umang Doshi explains why new methodologies are required to deal with local variation at advanced nodes.

Siemens’ Dan Scott looks at how the complexities of autonomy and electrification are driving significant changes to vehicle E/E architectures.

Cadence’s Paul McLellan digs into how the process of planting and harvesting on large farms is now assisted by highly computerized machines with GPS and automatic steering.

Synopsys’ Rahul Deokar explains why process variability, physical effects, and the impact of interconnect are critical in timing analysis.

FormFactor’s Peter Andrews suggests ways to add automation to a test setup without compromising low-noise performance.

YieldHub’s Marie Ryan offers reasons to purchase a yield analysis system instead of building it yourself.



Leave a Reply


(Note: This name will be displayed publicly)