Xiaomi EVs; Tesla Autopilot investigation; EDA anti-piracy measures.
Automotive
Chinese smartphone maker Xiaomi registered its electric vehicle (EV) business (Xiaomi EV) with an initial investment of RMB $10 billion (USD$1.5B). Xiaomi plans to invest USD $10 billion in Xiaomi EV over 10 years, under the leadership of Lei Jun, the founder of Xiaomi. Currently, Xiaomi EV has approximately 300 employees. Xiaomi also acquired autonomous driving technology company Deepmotion Tech last week.
Chinese EV companies Nio and Xpeng reported that they delivered fewer cars than expected in August because of chip shortages. U.S. EV maker Tesla is delaying its Roadster EV until 2023 because of supply chain and parts shortages. Other car companies — Ford, Toyota — are experiencing production shutdowns or slowdowns because of chip shortages.
U.S. EV truck and SUV startup Rivian filed an S1 form with the U.S. Securities and Exchange Commission for an IPO last week. Rivian has yet to release a product, but has orders already. Amazon and Ford are backing the startup.
A Tesla using its Autopilot self-driving automated driving module hit a Florida Highway Patrol cruiser. Injuries were minor. Regardless, Tesla has to turn over data from the incident to the U.S. Department of Transportation’s National Highway Traffic and Safety Administration (NHTSA), which will investigate for safety defects in the Autopilot. Tesla’s Autopilot was already the subject of NHTSA investigation to determine why the Autopilot has collided with parked first responder vehicles in the past. “Most incidents took place after dark and the crash scenes encountered included scene control measures such as first responder vehicle lights, flares, an illuminated arrow board, and road cones. The involved subject vehicles were all confirmed to have been engaged in either Autopilot or Traffic Aware Cruise Control during the approach to the crashes,” according to the NHTSA’s failure report.
Xilinx and automotive AI company Motovis are working together on autonomous vehicle systems for automotive forward facing cameras using Xilinx’s Automotive (XA) Zynq system-on-chip (SoC) platform and Motovis’ convolutional neural network (CNN) IP. According to a press release, OEMs and Tier-1 can now layer their own feature algorithms on top of the Motovis’ perception stack. The forward cameras help with lane-keeping assistance (LKA), automatic emergency braking (AEB), and adaptive cruise control (ACC), among other uses. Image resolution is up to eight megapixels on the systems.
Pervasive computing
Rambus demonstrated its PCIe 5.0 Digital Controller IP for FPGA platforms. Using a soft controller, the PCIe 5.0 performance runs at 32 GT/s data rates in FPGAs, which could enable use in multi-instance, switching and bridging applications. The controller IP helps accelerate FPGAs used in defense, networking — data centers, and test and measurement markets that need
Ansys is acquiring Zemax, whose high-performance optical imaging system simulation will add strengthen Ansys’ optical and photonic systems modeling portfolio. Optical systems are used in robotic surgery-assist systems, autonomous vehicles, machine vision cameras. and aerospace and defense applications, says a press release from Ansys. The acquisition is targeted to close in the fourth quarter of 2021, subject to regulatory approvals. A Definitive Acquisition Agreement exists between Zemax’ owner EQT Private Equity and Ansys.
The MIPI Alliance updated C-PHY and D-PHY. The MIPI C-PHY update — version 2.1— has a new 64-bit PHY Protocol Interface (PPI) for a wider bus between C-PHY and a chip’s core logic for higher performance. C-PHY now supports symbol rates up to 6 Gigasymbols per second (Gsps) — which equals 13.7 Gbps — over the standard channel and up to 8 Gsps over the short channel, according to a MIPI press release. The D-PHY update — version 3.0 — adds a Continuous-Time Linear Equalizer (CTLE) on the receiver side that doubled the data rate of D-PHY’s standard channel to 9 Gigabits per second (Gbps). D-PHY is physical-layer interface that first used in cameras and smartphone displays but now has broadened to similar uses in other devices, even automotive.
China Mobile, Huawei, and 10 other participants formed a 5G VoNR+ (New Calling) Working Group. Voice over 5G New Radio is a basic call service that uses the SA architecture of 5G network. Ericsson and Vodafone were able to cut power use by 43% in a 5G New Radio test installation in London, using Ericsson’s antenna-integrated radio solution (AIR 3227). SkyWater will produce solid-state holographic optical beam-steering chips for Carillon Technologies as part of a DARPA-funded Satellite Communications Program. The goal is to creating lightweight, low-cost satellite-to-satellite laser communication links for satellite and other free-space optical communications (FSOC) applications. The solid-state chip will replace moving mirrors.
Marvell is buying Innovium, a switch chip maker, for $1.1 billion. Marvell plans to co-package optics from its Inphi acquisition alongside the high-bandwidth, low-latency switch chips. Innovium’s Teralynx chip family is used in large-scale data centers.
Security
SEMI’s Electronics System Design (ESD) Alliance has completed an anti-piracy Server Certification Protocol (SSCP) that will help keep unauthorized users from accessing a semiconductor design software license. The standard will ensure that only authorized servers are issuing licenses. Cadence, Siemens EDA, and Synopsys all participated in the development and signed off on SSCP. Next step for SSCP is create an industry standard based on the protocol, after which the SEMI Standards group will take over managing the standard.
Bitdefender found vulnerabilities in a baby monitor’s camera that allowed unauthorized access to data and remote control of the camera. In the Victure PC420 smart camera, a local stack-based buffer overflow where remote code can be executed is described in CVE-2020-15744.
Cybersecurity firm Fireeye with CISA disclosed a flaw on August 17 that could affect millions of IoT devices that use ThroughTek’s Kalay network. The vulnerability enables attackers to access the IoT network to pull data from and control other devices on it. The malware resides in memory and is a fileless, which makes it different from other malware, writes Dark Reading’s Jai Vijayan. The vulnerability is tracked under CVE-2021-28372.
Read the most recent Automotive, Security, & Pervasive Computing newsletter. Check out 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