Millimeter waves for automotive communications and radar

Millimeter waves for automotive communications and radar

Technology News |
By eeNews Europe

Frequencies between 10 GHz and 30 GHz have been used in cellular research by NYU Wireless, whose director is a former UT Austin professor. While NYU Wireless studied mmW propagation in New York City, the UTA Wireless Networking and Communications team lead by Professor Robert Heath Jr. is trying to predict mmW performance.

Heath said UTA faculty and students have taken the insights from NYU Wireless and developed mathematical performance models for mmW and massive MIMO (multiple-input multiple-output), a technique for using a large number of antennas seen as key to future cellular base stations.

Millimeter waves work well when densely deployed in areas that don’t have much blockage, Heath told EE Times. Massive MIMO on a low frequency could overlay a whole city, then small areas could be lit up with mmW for additional cellular coverage.

Heath expanded upon this idea for use in connected or automated vehicles.

Partially automated car UT vision for the future of transportation. Source: University of Texas, Austin.

UTA leveraged the mmW consumer WLAN standard 802.11ad to develop a vehicular communication-radar waveform for long range automotive radar (LRR) and vehicle-to-vehicle communication (V2V) at 60 GHz. Automotive radar operates at 77 GHz, making the bands close together enough to do radar on mmW.

Communication is lagging tremendously behind everything [in connected cars]. We have radars on cars, cameras, interest in IR, and yet communications is through the DSRC [dedicated short-range communications] standard or cellular. The problem is all sensors generate a huge amount of raw data. I think car companies are going to want to get access to sensor data on other vehicles….I think there’s going to be interest in streaming high data low latency data from cars.

Heath’s team of 15 researchers have modeled how mmW could serve as the framework for V2V and vehicle-to-infrastructure (V2I) communications for congestion or collision reduction. The model shows both the effects of beam width and pointing error (the ease with which a narrow beam is pointed at an incorrect location) to show how signals vary over time. The team found that directional antennas with an approximately 3 degree width support stable channels when well pointed.

Millimeter waves could be beamed from atop a light pole (above), though building communications into infrastructure poses its own challenges. Source: University of Texas, Austin

Millimeter waves could be beamed from atop a light pole (above), though building communications into infrastructure poses its own challenges. Source: University of Texas, Austin.

“There are a lot of claims made that millimeter waves are not suitable for use in vehicles because they move too fast and the wavelength is too small,” Heath said, adding that higher frequencies propagate faster, making communications difficult. “If you point a very narrow beam these effects go away. The narrower it is, the longer the channel takes to change.”

Heath will present the model at IEEE’s vehicular conference this fall. He hopes the mathematical models will help inform engineers how many antennas to use at a given time.

To further improve communications, Heath’s group teamed up with UT Austin’s Center for Transportation Research to develop a mmW and radar communication prototype that does V2V and V2I. The prototype will use a National Instruments PXI chassis interface with custom RF, and aims to leverage the close location of radar and 60 GHz frequencies to cut down on hardware cost.

mmW communication and radar prototype. Source: University of Texas, Austin

A mmW communication and radar prototype. Source: University of Texas, Austin.

UTA’s Wireless Networking and Communications Group is also working on ways to use 802.11 OFDM waveforms on a single forward directional antenna at 5.9 GHz for collision detection.

“The problem with radar is there are a lot of ways to fake the signal, something that causes a processing engine to take evasive action. It has many potential insecurities,” Heath said. “We’re looking at a couple ways to augment radar with communications.”

Mimicking radar, the OFDM wave communicates with a car in front through two antennas while simultaneously listening to that car’s echo. Heath said this communication method could make many message types possible and, with some signal processing, could allow for better accuracy than what other bandwidths would dictate.

Most new cars likely will have radar as a standard feature in two years, Heath predicted – and if those cars also have an embedded communications chip, information from the vehicle can be cross-validated with existing hardware.

Heath’s team is working on a vehicle security prototype that uses 802.11p at 5.8 GHz to communicate alongside a 77 GHz radar and a custom wave form generator from National Instruments “to show that communication helps you avoid the spoofing problem, and improve radar security.”

Research into the overlap between radar and communications is sponsored by the Texas Department of Transportation with the goal of using joint sensor data through machine learning and sensor fusion. Heath’s team hopes to collaborate further on radar security projects and prototyping.

The author, Jessica Lipsky is Associate Editor of EE Times –

If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :    eeNews on Google News


Linked Articles