4 ms·

Integrated sensing and communication based on space-time-coding metasurfaces

40 points by westurner 2y ago

westurner 2y ago
ScholarlyArticle: "Integrated sensing and communication based on space-time-coding metasurfaces" (2025) https://www.nature.com/articles/s41467-025-57137-6 https://www.nature.com/articles/s41467-025-57137-6 NewsArticle: "Space-time-coding metasurface could transform wireless networks with dual-functionality for 6G era" https://techxplore.com/news/2025-03-space-coding-metasurface-wireless-networks.html https://techxplore.com/news/2025-03-space-coding-metasurface...
cr125rider 2y ago
So this is a sort of wireless amplifier?
- RossBencina 2y ago
  I think it's more like an active antennae/analog signal processing structure that can tightly control spatial radiation properties (direction, amplitude, phase). Think beamforming, but with a continuous micro-structured surface rather than discrete antennae. In this case, the "antennae" is not only a planar object with sub-wavelength microstructure, but may also include circuit elements that can dynamically alter the surface properties. At least that's what I got from skimming the first three paragraphs. Sounds very interesting.
  - exe34 2y ago
    I love this sort of thing, it reminds me of Neal Asher's polity enemies who often have devices being grown on their surface where powerful laser beams or other weapons emerge.
mikewarot 2y ago
This seems like a wild work around to retrofit a non-MIMO system into the modern era with beamforming chips a-plenty. While it's interesting to consider using an AI to detect and steer the beam using the harmonics of the received signal, it's important to bear in mind these will be about 20 dB down from the fundamental received frequency. It would be far more reliable to employ a phased array transceiver which can do multiple concurrent beamforming operations. This would allow the use of the antenna for multiple connections on the fundamental frequency. If it's truly for a single channel/single user, it would be simpler to simply time-multiplex a search for reflections by random sampling, or perhaps even gradient descent.
ydjje 2y ago
[dead]

40 points by westurner 2y ago

ScholarlyArticle: "Integrated sensing and communication based on space-time-coding metasurfaces" (2025) https://www.nature.com/articles/s41467-025-57137-6 https://www.nature.com/articles/s41467-025-57137-6 NewsArticle: "Space-time-coding metasurface could transform wireless networks with dual-functionality for 6G era" https://techxplore.com/news/2025-03-space-coding-metasurface-wireless-networks.html https://techxplore.com/news/2025-03-space-coding-metasurface...

cr125rider 2y ago

So this is a sort of wireless amplifier?

RossBencina 2y ago
I think it's more like an active antennae/analog signal processing structure that can tightly control spatial radiation properties (direction, amplitude, phase). Think beamforming, but with a continuous micro-structured surface rather than discrete antennae. In this case, the "antennae" is not only a planar object with sub-wavelength microstructure, but may also include circuit elements that can dynamically alter the surface properties. At least that's what I got from skimming the first three paragraphs. Sounds very interesting.
- exe34 2y ago
  I love this sort of thing, it reminds me of Neal Asher's polity enemies who often have devices being grown on their surface where powerful laser beams or other weapons emerge.

mikewarot 2y ago

This seems like a wild work around to retrofit a non-MIMO system into the modern era with beamforming chips a-plenty. While it's interesting to consider using an AI to detect and steer the beam using the harmonics of the received signal, it's important to bear in mind these will be about 20 dB down from the fundamental received frequency. It would be far more reliable to employ a phased array transceiver which can do multiple concurrent beamforming operations. This would allow the use of the antenna for multiple connections on the fundamental frequency. If it's truly for a single channel/single user, it would be simpler to simply time-multiplex a search for reflections by random sampling, or perhaps even gradient descent.

ydjje 2y ago

[dead]