Week In Review: Auto, Security, Pervasive Computing

DSPs on AI edge; medical IoT more insecure; Toyota HD maps.


AI, machine learning
Cadence says it has optimized its Tensilica HiFi digital signal processor IP to efficiently execute TensorFlow Lite for Microcontrollers, which are used in Google’s machine learning platform for edge. This means developers of AI/ML on the edge systems can now put better audio processing on edge devices with ML applications like keyword detection, audio scene detection, noise reduction and voice recognition.

Synopsys updated its Simpleware ScanIP software by adding machine learning that can orthopedic surgeons prepare for surgery more quickly. The ML-based Simplware AS Ortho module gives medical staff who are planning hip and knee surgeries an automated processing of 3D images of a patient’s anatomy. “Image segmentation of MRI and CT scans presents a significant challenge for our surgical and engineering multidisciplinary teams. We’re excited to collaborate with the Simpleware product group at Synopsys for solutions to this challenge,” said Johann Henckel MD, Orthopedic Surgeon, Royal National Orthopaedic Hospital, UK, in a press release. “What is currently a laborious process that occupies significant engineering resources and time can now be completed quickly, accurately and with less variability, promising a scalable solution for generating high-fidelity patient specific models, surgical tools and bespoke implants.” Other modules for specific parts of human anatomy will be added later.

Synopsys also announced DSO.ai, which uses artificial intelligence on chip design. According to its press release, an artificial intelligence and reasoning engine capable of searching for optimization targets in very large solution spaces of chip design.

A CMOS chip may help detect coronavirus (COVID-19) faster. The company Anitoa has introduced a rapid, onsite nucleic acid test for optical detection of coronavirus in 30 minutes, reports BioOptics World. The device uses a CMOS biosensor chip that Anitoa, a Chinese-based company, developed and has a 97% accuracy rate according to the company. With other tests, samples have to be packaged, frozen, sent to a lab, thawed before testing begins. The COVID-19 lab tests are PCR tests, according to Oxford University’s Our World In Data, and so far South Korea is the best at testing the most people.

The United States Food and Drug Administration also approved a COVID-19 test from Roche that takes 10 minutes to complete the test, but is still a larger machine. The Roche cobas 6800 / 8800 systems were approved. The 8800 system is supposed to be able to test 4,128 patients a day, the 6800 version can test  1,440, according to a Bloomberg article.

Edge, data center, cloud
Cadence is working with STMicroelectronics to tape out a 56G very short-reach (VSR) SerDes in 7nm for a system on chip (SoC) targeted at the networking, cloud and data center markets.

Over 83% of medical imaging devices are running on unsupported operating systems, says Palo Alto Networks’ Unit 42, in a new IoT Threat Report. This number jumped 56% in 2019 because of Windows 7 operating system is not being updated anymore. The report says the IoT medical devices with the most security issues are imaging systems.

Another threat arrived in a different form. Two attendees at last month’s RSA Conference in San Francisco contracted the coronavirus. RSA is one of the largest security conferences.

A Beijing-based driverless delivery business has booked orders for more than 200 vehicles in the past two months, reports Bloomberg. Startup Neolix’s inventories have been depleted during the COVID-19 epidemic. The vehicles are used to deliver supplies to hospitals and its vans are used for disinfecting the streets and food delivery.

Toyota has demonstrated a high-definition map building system for automated driving using satellite imagery and data from regular car cameras with a relative accuracy of 50 cm (20 inches). Data from special survey vehicles are not needed, and the system can automatically clean up data using algorithms to adapt to different vendor systems. Working with earth intelligence and space infrastructure company Maxar Technologies, NTT DATA Corporation and data platform company HERE Technologies, Toyota Research Institute-Advanced Development, Inc. (TRI-AD) has developed its Automated Mapping Platform (AMP). The AMP can automatically remove automobiles that show up on roads in satellite imagery and generate lane map that can be updated in real time on a TomTom’s HD map. TRI-AD is Toyota’s automated driving software development company.

Qualcomm announced that many Cellular Vehicle-to-Everything (C-V2X) products with its chipset, Qualcomm’s 9150 C-V2X, have achieved the certification needed to go on the European market. The European Radio Equipment Directive (RED) Certification is required for C-V2X products in Europe. C-V2X is protocol used for cars to communicate with each other and edge devices mounted on infrastructure such as light poles and buildings.

Silicon photonics, which gaining momentum in automotive applications, has an improved die and wafer testing tool. FormFactor’s silicon photonics (SiPh) CM300xi probing system announced new features that make it possible to measure performance more realistically by probe couple fibers at the edge of SiPh devices on die and wafers. The system can also measure power on single fibers and fiber arrays to examine insertion path loss. CM300xi works with Keysight‘s Test Automation Platform software.

Shared control between driver and car is the key, says the IIHS (Insurance Institute for Highway Safety) safety recommendations on the design of partially automated driving systems, issued this week. The guidelines emphasize how to keep drivers focused on the road even as the vehicle does more of the work. According ot IIHS, Research has shown that the more sophisticated and reliable automation becomes, the harder it is for a driver to remain vigilant.

Job, Event and Webinar Boards: Find industry jobs and upcoming conferences and webinars all in one place on Semiconductor Engineering. Knowledge Center: Boost your semiconductor industry knowledge. Videos: See the latest Semiconductor Engineering videos.

Leave a Reply

(Note: This name will be displayed publicly)