Through-wall-imaging (TWI) technologies do just what their name suggests: Allow users to “see” through walls. Most radar offerings, however, call for a range of frequencies – a broad capability that increases costs.
Current TWI systems also require advanced knowledge of the wall’s materials; software frequently predicts how a given barrier will affect the scanning wave, separating echoes and distortions from the solid objects being sought.
Using a narrow band of microwave frequencies, researchers at Duke University found a way to see on the other side of the wall, regardless of the divider’s makeup. Instead of requiring large-bandwidth radar to separate the object from its surroundings, Duke’s technology monitors a different wall feature: symmetry.
By accounting for the types of distortions typically created by flat, uniform walls, the new algorithm enables higher-resolution scans.
“We wrote an algorithm that separates the data into parts — one that shows circular symmetry and another that doesn’t,” said Okan Yurduseven, a postdoctoral researcher in electrical and computer engineering at Duke. “The data that doesn’t have any symmetry is what we’re trying to see.”
We at Tech Briefs spoke with Daniel Marks, associate research professor of electrical and computer engineering at the university, who led the development of the technology . In the edited Q&A below, Marks takes us through the new applications he envisions for an upgraded TWI.
Tech Briefs: What are the biggest technical challenges when trying to “see through walls?”
Daniel Marks: There are three essential challenges to seeing through walls. The first is to obtain a signal from the objects behind the wall, which is typically achieved by selecting electromagnetic frequencies that can penetrate the wall and scatter from the objects.
The second is to separate the objects’ signal from confounding signals, such as the signal from the wall itself.
Finally, the image must be formed from the signal. The materials the wall and objects consist of largely determine which frequency may be used.
Tech Briefs: What are the weaknesses of current TWI options?
Marks: Current through-wall-imaging radars typically use a high bandwidth signal so that the object and the wall can be separated; the signal penetrating through the wall takes longer to return to the radar system than the signal that reflects directly from the wall. In order to distinguish objects that are separated by only a centimeter or two, however, a very large bandwidth radar is needed, which increases cost and makes certification by telecommunications agencies more difficult.
Furthermore, even if the signals are separated by their time of arrival, the distortion produced by the wall is unknown, and so must be determined by other means for a clear image to be formed.
Tech Briefs: How does your technology improve upon these previous offerings?
Marks: In general, walls are typically constructed from layered materials. Layered materials produce circularly symmetric distortions to the signal captured from behind the wall, while the objects behind the wall are usually not symmetric. By separating the return signal into a symmetric and asymmetric part, the symmetric component may be attributed to the wall, and the distortions it causes and the asymmetric part are attributed to the objects behind it.
Therefore, a large bandwidth radar is not needed to separate the object and the wall, as these are distinguished by symmetry instead. Furthermore, this automatically solves the problem of forming a clear image, as the symmetric distortions due to the wall are removed from the asymmetric component of the signal, leaving the undistorted asymmetric signal from the object.
Tech Briefs: How is the system able to operate without advance knowledge of the wall material?
Marks: The layered structure produces a circularly symmetric distortion to the return signal which is circularly symmetric regardless of the wall material (as long as the wall is built as a layered structure). Therefore, the distortion due to any layered medium placed in front of the object may be inferred from the return signal, as long as enough return signal is obtained from the object.
Tech Briefs: Take us through an application that you envision.
Marks: The two main application areas we envision are for construction and security. For construction, it is helpful to visualize the infrastructure embedded in walls and ceilings without having to breach the wall. A radar sensor placed on a smartphone could capture the signal scattered from the wall as it is moved, say, in a spiral pattern.
Based on this data, the image would be formed, removing the symmetric distortions and showing the results as an image of the signal strength returned from the objects behind the wall. For real-time imaging for security applications, multiple sensors could be employed together to decrease the time required to capture an image.
Tech Briefs: What is most exciting to you about your TWI system?
Marks: Because separating signals by their symmetry in an imaging problem can achieve remarkably good results and simplify and economize the hardware, this approach is applicable to many transceiver systems — even those not even specifically designed for imaging, such as Wi-Fi and mobile phone radios. These other devices may be used to image as well, opening up the opportunity to form images from the signals gathered from all of these devices that are ubiquitous in the environment.
Tech Briefs: What’s next regarding the technology development?
Marks: Combining the radar with machine vision systems is a near-term goal that will enable these radar images to be superimposed on the camera images, as well as help register these images in the environment so that the through-wall data of a building can be archived in a manner to how street-view image databases are already used, but for interior spaces.
This work was supported by the Air Force Office of Scientific Research (FA9550-12-1-0491).
What do you think? Share your thoughts below.
- The Cassini Radio Detection and Ranging (RADAR) will investigate the surface of Titan.
- Read more Imaging tech briefs.