Automotive Rising! GLOBALFOUNDRIES’ 22FDX® (FD-SOI) Accelerates

By: Rajeev Rajan

There’s no doubt about it, the ‘best of tech’ was on display at this year’s Mobile World Congress, where top manufacturers gathered in Barcelona, Spain to showcase their latest products. Considered the Mobile event, the showroom floor was flooded with the world’s coolest gadgets and devices, some far from being in (mobile) phone form-factor. Thanks to increasingly advanced chipsets and emerging 5G technologies,  which are set to open a new era of connectivity, compelling new technologies are being showcased – from wearables, to VR cameras, to autonomous cars – at major tech events such as CES and MWC.

Among the new tech, were products for the connected car – not something you’d typically see at this venue, but some of the biggest players in mobile are now taking interest in the connected car space, and showcasing a connected car product right alongside its new flagship smartphone. And, the brains behind the devices in these ‘smart’ cars are semiconductors. Semiconductors are rising in auto-tech and carmakers are increasingly trying to find ways to integrate consumer electronics functions such as Advanced Driver Assistance Systems, or ADAS, applications and driver-assistance technologies into their vehicles. Capabilities such as vision processing, cameras, sensors, connectivity, and mapping take ADAS and autonomous driving to a new level.

A case in point, Dream Chip Technologies announcement of the industry`s first 22nm FD-SOI silicon for a new ADAS System-on-Chip (SoC) for automotive computer vision applications during MWC Barcelona.

Now, ADAS SoCs have been around for more than 5 years. So what’s the big deal, you wonder? Dream Chip’s multi-processor SoC solution – developed under the European THINGS2DO project in coordination with ARM, Cadence, INVECAS, Arteris, and GF – is the industry’s first SoC designed on GF’s 22FDX® (FD-SOI) process technology. And, this is important if you’re a designer trying to figure out how to reduce complexity while watching your bottom line at the same time.

As a complementary path to FinFETs, GF’s 22FDX transistors exhibit Superior RF/Analog – and a high fT and Fmax (350/325GHz), making it an ideal technology for low-power 5G and mmWave applications. Also, the technology platform’s back-biasing capability plays a role in controlling power by raising the threshold voltage and reducing leakage where appropriate, making it a power-efficient fit for wireless, battery-powered computing applications. And, what about cost? With 22nm design rules, single patterning is sufficient, eliminating double patterning and providing a cost-competitive product.

Beyond the design process, there are other ‘bells and whistles’ the optimized SoC provides such as video input and output interfaces as well as common communication interfaces on two board-to-board headers. And, some of the key ADAS processing use-cases that can be realized with such an SoC include 360 degree top view, road-sign recognition, lane departure warning, driver distraction warning, blind spot detection, surround vision, flicker mitigation for digital mirroring, pedestrian detection, cruise control and emergency braking.

In addition, Dream Chip’s SoC design incorporates multiple IPs such as foundation IPs, LPDDR4, PLL, Thermal Sensor and Process Monitor, from INVECAS as well as two LPDDR4 3200 high bandwidth memory interfaces.

Dream Chip’s ADAS SoC, with GF’s 22FDX process tech, is a prime example of the brain power behind the ‘rise’ of semiconductor technologies that are fueling the growth of automotive and other global markets. These semiconductor devices are elevating consumer experiences that demand simpler, smaller, faster and increasingly lower cost end products. GF technologies, products, and solutions are differentiating key applications in these end spaces.

If you missed the buzz of the Dream Chip automotive SoC at Mobile World Congress, we invite you to contact a GF sales representative or visit Dream Chip’s web site at: www.dreamchip.de to learn more.