Forum for Science, Industry and Business

Sponsored by:     3M 
Search our Site:

 

Machine learning shapes microwaves for a computer's eyes

10.01.2020

Hardware and software tweak microwave patterns to discover the most efficient way to identify objects

Engineers from Duke University and the Institut de Physique de Nice in France have developed a new method to identify objects using microwaves that improves accuracy while reducing the associated computing time and power requirements.


An example of a wave pattern (right) and its intensity levels (left) developed by the machine learning algorithm to best illuminate the most important features of an object being identified.

Credit: Mohammadreza Imani, Duke University


In a new type of object identification, a radio wave source (back panel) creates a wave front (middle panel) that is shaped by a metamaterial screen which allows waves to pass through in some places but not others (front panel). Machine learning then finds the wave shapes that illuminate the most useful features of an object. The method improves accuracy while reducing computing time and power requirements.

Credit: Mohammadreza Imani, Duke University

The system could provide a boost to object identification and speed in fields where both are critical, such as autonomous vehicles, security screening and motion sensing.

The new machine-learning approach cuts out the middleman, skipping the step of creating an image for analysis by a human and instead analyzes the pure data directly. It also jointly determines optimal hardware settings that reveal the most important data while simultaneously discovering what the most important data actually is.

In a proof-of-principle study, the setup correctly identified a set of 3D numbers using tens of measurements instead of the hundreds or thousands typically required.

The results appear online on December 6 in the journal Advanced Science and are a collaboration between David R. Smith, the James B. Duke Distinguished Professor of Electrical and Computer Engineering at Duke, and Roarke Horstmeyer, assistant professor of biomedical engineering at Duke.

"Object identification schemes typically take measurements and go to all this trouble to make an image for people to look at and appreciate," said Horstmeyer. "But that's inefficient because the computer doesn't need to 'look' at an image at all."

"This approach circumvents that step and allows the program to capture details that an image-forming process might miss while ignoring other details of the scene that it doesn't need," added Aaron Diebold, a research assistant in Smith's lab. "We're basically trying to see the object directly from the eyes of the machine."

In the study, the researchers use a metamaterial antenna that can sculpt a microwave wave front into many different shapes. In this case, the metamaterial is an 8x8 grid of squares, each of which contains electronic structures that allow it to be dynamically tuned to either block or transmit microwaves.

For each measurement, the intelligent sensor selects a handful of squares to let microwaves pass through. This creates a unique microwave pattern, which bounces off the object to be recognized and returns to another similar metamaterial antenna. The sensing antenna also uses a pattern of active squares to add further options to shape the reflected waves. The computer then analyzes the incoming signal and attempts to identify the object.

By repeating this process thousands of times for different variations, the machine learning algorithm eventually discovers which pieces of information are the most important as well as which settings on both the sending and receiving antennas are the best at gathering them.

"The transmitter and receiver act together and are designed together by the machine learning algorithm," said Mohammadreza Imani, research assistant in Smith's lab. "They are jointly designed and optimized to capture the features relevant to the task at hand."

"If you know your task, and you know what sort of scene to expect, you may not need to capture all the information possible," said Philipp del Hougne, a postdoctoral fellow at the Institut de Physique de Nice. "This co-design of measurement and processing allows us to make use of all the a priori knowledge that we have about the task, scene and measurement constraints to optimize the entire sensing process."

After training, the machine learning algorithm landed on a small group of settings that could help it separate the data's wheat from the chaff, cutting down on the number of measurements, time and computational power it needs. Instead of the hundreds or even thousands of measurements typically required by traditional microwave imaging systems, it could see the object in less than 10 measurements.

Whether or not this level of improvement would scale up to more complicated sensing applications is an open question. But the researchers are already trying to use their new concept to optimize hand-motion and gesture recognition for next-generation computer interfaces. There are plenty of other domains where improvements in microwave sensing are needed, and the small size, low cost and easy manufacturability of these types of metamaterials make them promising candidates for future devices.

"Microwaves are ideal for applications like concealed threat detection, identifying objects on the road for driverless cars or monitoring for emergencies in assisted-living facilities," said del Hougne. "When you think about all of these applications, you need the sensing to be as quick as possible, so we hope our approach will prove useful in making these ideas reliable realities."

###

This research was supported by the Air Force Office of Scientific Research (FA9550-18-1-0187).

CITATION: "Learned Integrated Sensing Pipeline: Reconfigurable Metasurface Transceivers as Trainable Physical Layer in an Artificial Neural Network." Philipp del Hougne, Mohammadreza F. Imani, Aaron V. Diebold, Roarke Horstmeyer, and David R. Smith. Adv. Sci. 2019, 1901913. DOI: 10.1002/advs.201901913

Media Contact

Ken Kingery
ken.kingery@duke.edu
919-660-8414

 @DukeU

http://www.duke.edu 

Ken Kingery | EurekAlert!
Further information:
https://pratt.duke.edu/about/news/machine-learning-object-identification
http://dx.doi.org/10.1002/advs.201901913

More articles from Information Technology:

nachricht Tiny optical cavity could make quantum networks possible
31.03.2020 | California Institute of Technology

nachricht Chip-based devices improve practicality of quantum-secured communication
23.03.2020 | The Optical Society

All articles from Information Technology >>>

The most recent press releases about innovation >>>

Die letzten 5 Focus-News des innovations-reports im Überblick:

Im Focus: Physicist from Hannover Develops New Photon Source for Tap-proof Communication

An international team with the participation of Prof. Dr. Michael Kues from the Cluster of Excellence PhoenixD at Leibniz University Hannover has developed a new method for generating quantum-entangled photons in a spectral range of light that was previously inaccessible. The discovery can make the encryption of satellite-based communications much more secure in the future.

A 15-member research team from the UK, Germany and Japan has developed a new method for generating and detecting quantum-entangled photons at a wavelength of...

Im Focus: Junior scientists at the University of Rostock invent a funnel for light

Together with their colleagues from the University of Würzburg, physicists from the group of Professor Alexander Szameit at the University of Rostock have devised a “funnel” for photons. Their discovery was recently published in the renowned journal Science and holds great promise for novel ultra-sensitive detectors as well as innovative applications in telecommunications and information processing.

The quantum-optical properties of light and its interaction with matter has fascinated the Rostock professor Alexander Szameit since College.

Im Focus: Stem Cells and Nerves Interact in Tissue Regeneration and Cancer Progression

Researchers at the University of Zurich show that different stem cell populations are innervated in distinct ways. Innervation may therefore be crucial for proper tissue regeneration. They also demonstrate that cancer stem cells likewise establish contacts with nerves. Targeting tumour innervation could thus lead to new cancer therapies.

Stem cells can generate a variety of specific tissues and are increasingly used for clinical applications such as the replacement of bone or cartilage....

Im Focus: Artificial solid fog material creates pleasant laser light

An international research team led by Kiel University develops an extremely porous material made of "white graphene" for new laser light applications

With a porosity of 99.99 %, it consists practically only of air, making it one of the lightest materials in the world: Aerobornitride is the name of the...

Im Focus: Cross-technology communication in the Internet of Things significantly simplified

Researchers at Graz University of Technology have developed a framework by which wireless devices with different radio technologies will be able to communicate directly with each other.

Whether networked vehicles that warn of traffic jams in real time, household appliances that can be operated remotely, "wearables" that monitor physical...

All Focus news of the innovation-report >>>

Anzeige

Anzeige

VideoLinks
Industry & Economy
Event News

“4th Hybrid Materials and Structures 2020” takes place over the internet

26.03.2020 | Event News

Most significant international Learning Analytics conference will take place – fully online

23.03.2020 | Event News

MOC2020: Fraunhofer IOF organises international micro-optics conference in Jena

03.03.2020 | Event News

 
Latest News

Phage capsid against influenza: Perfectly fitting inhibitor prevents viral infection

31.03.2020 | Life Sciences

A 'cardiac patch with bioink' developed to repair heart

31.03.2020 | Life Sciences

Artificial intelligence can speed up the detection of stroke

31.03.2020 | Medical Engineering

VideoLinks
Science & Research
Overview of more VideoLinks >>>