Enabling cars that truly sense their environment
Imec, the European nanotech centre in Leuven (Belgium), recently made the world's first 79GHz radar transmitter in plain digital 28nm CMOS with an output power above 10dBm. In non-technical speak, that is the first high-resolution radar chip for the masses.
There are already radar chips out there. But they are fabricated in low volumes and in dedicated technologies that make them quite expensive. They are mainly used in military applications, and are beginning to find their way into more expensive vehicles.
Now, with this news, we may see the radars for the masses. Radars that are made in todays mainstream chip technology. That can be mass-produced at a low cost. Radars that are so small you could easily integrate them invisibly. In a car, but also in a bike helmet, a detection camera, or infrastructure such as automatic lights or door openers.
The most disrupting application however might lie in the field of robotics. Autonomous drones, both aerial and ground-based, can be deployed to deliver parcels, automate warehousing, to clean and maintain, for inspection and surveillance. High-resolution millimeter-wave radar would enable truly autonomous navigation for these robots in all circumstances.
What is a radar chip?
In your smartphone, there is a set of chips and antennae that take care of the wireless communication. They send out information as a radio wave, and conversely capture radio waves of a certain frequency and convert these back into digital information.
These same electromagnetic waves that we use for communication can also make a radar. The signal that is transmitted is always partially reflected on the surrounding objects. If we can receive that echo and compute the distance the signal has travelled, we have a radar: a means to sense the objects around us.
Wireless communication is widespread, very accessible, and wildly popular. That is partly because manufacturers have agreed on standards, so that all kinds of appliances can communicate. The same is now happening for radar technology. There is e.g. a growing industry consensus to use the 79GHz frequency band.
Why use high-frequency 79GHz signals?
79GHz waves, and mm-waves in general, are special for two reasons. First, they are very robust, allowing to see in darkness, fog and snow. And second, they allow to see more details than lower-frequency signals.
All waves that travel through the atmosphere are attenuated. Mm-waves are attenuated even more than visible light. So why use them? Because they fare better in fog, snow, and heavy rain. These almost totally block the visual light, while mm-waves get through unharmed. So high-frequency radio waves see where we cannot.
Second, mm-waves give a better resolution. Designing a chip that covers 5% of the carrier frequency, is more or less equally difficult at 2GHz as at 79GHz. But the result is that you have 4GHz of bandwidth at 79GHz instead of 100MHz at 2GHz. In communication, more bandwidth results in more data. For a radar, you get a better depth resolution.
Why would we need those radars?
Safety in the first place. As a car driver, even an attentive driver, you cannot look at the complete environment of the car. You are focused on a narrow zone in front of you, making you vulnerable for what happens outside that zone. Moreover, an electronic system can react much faster than we can. And if for whatever reason, your attention is waning, then you really could use some help.
But once a radar chips is cheap enough, readily available, and really small, it could also be integrated in a bike helmet (or skate helmet, ski helmet ...). Warning kids for upcoming cars, playing toddlers, or other obstacles that could harm them.
And instead of the current movement detectors in houses, we could use more intelligent radars. Radars that are able to see and compute who is moving (a cat or a thief), and what the movement is. That way, you could even think of devices that are sensitive to specific movements. Did anyone say 'gaming?
About the author
Jan Provoost is science editor at imec, reporting about imecs research and R&D results in international magazines and newsletters. Jan has a Masters degree in Languages (1989) and in Computer Science (1993), both from KU Leuven. Before joining imec, he worked as researcher and editor for various high-tech start-ups.
To download the PDF version of this article, click here.
|Related Articles||Editor's Choice|