Unlike WiFi or cellular signals, millimeter wave wireless communications channels are unable to pass through walls, furniture or other objects. This reality is part of what is holding back implementation of millimeter wave wireless, which is of great interest for 6G wireless systems and beyond because of its extremely large data-carrying capacity.
Electrical engineers from the University of California San Diego have developed and demonstrated an inexpensive solution to this millimeter wave wireless roadblock — and in a paper presented on August 19 at ACM SIGCOMM 2026, they report that their passive solution performs as well or better than today's expensive workarounds to the path-obstruction problem.
The team designed and produced 3D-printed tiles that reflect millimeter wave signals in precise ways based on a map of the indoor environment that needs wireless coverage and the location of the hardware sending and receiving the signals. Using this strategy, the UC San Diego engineers demonstrate the ability to:
Nearly double average millimeter wave communications link data rates indoors. (The engineers say their work is relevant for outdoor environments as well.)
Double the wireless coverage area in challenging indoor environments.
These performance gains match those that can be achieved with active reconfigurable intelligent surfaces (RIS) — a leading candidate technology in current 6G research for addressing the path-obstruction problem — but at a fraction of the cost and complexity.
This new work is from a team of engineers led by Xinyu Zhang, a professor of electrical and computer engineering at UC San Diego and Director of the UC San Diego Center for Wireless Communications. Zhang is the senior author on the paper presented at SIGCOMM, one of the world's premier research conferences in computer networking.
Passive Reflective Tiles
Each tile is lightweight, inexpensive and easy to adhere to wall surfaces. A six inch by six inch tile contains many thousands of elements smaller than the millimeter waves themselves. These sub-wavelength units passively control the flow of the wireless signals that hit the tile.
In broad strokes, the researchers developed two classes of reflective tiles: tiles made to move larger data streams throughout the indoor environment, and tiles that spread the signal into areas where it is available for use by end users. They call their system FlowForm.
Using these reflective tiles, the engineers mapped five different indoor environments and then designed site-specific sets of tiles in order to ensure that the millimeter waves are widely distributed within the environment, with redundancies needed to get around corners, walls and even people.
The researchers note their passive approach is compatible with all relevant millimeter wave protocols.
"Millimeter wave wireless technologies have a lot of potential given the incredible amount of data they can carry. Our work demonstrates that a collective set of passive surfaces can make millimeter wave networks robust in real-world environments, with significantly lower complexity and cost compared with mainstream active reconfigurable intelligent surfaces," said Zhang, the Ericsson Endowed Chair Professor at UC San Diego. He is on the faculty in the Department of Electrical and Computer Engineering at the UC San Diego Jacobs School of Engineering.
FlowForm
At ACM SIGCOMM 2026, the UC San Diego electrical engineers are presenting their radically different approach to the millimeter wave wireless path-obstruction problem. Their FlowForm approach replaces active infrastructure with coordinated networks of entirely passive metasurfaces.
The team's 3D-printed reflective tiles are made of plastic filament with a conductive paint layer. They cost roughly $2 each and require no power, no control, and no runtime coordination.
The key insight is that organizing many cheap passive surfaces into a coordinated network can rival the performance of active systems.
FlowForm introduces a hierarchical design approach inspired by how smaller tributaries feed into rivers — but in reverse. The engineers call their approach the major–minor flow topology. Major flows are focused relay chains that shuttle millimeter wave wireless signals across long distances and around obstacles, forming the communication backbone. Minor flows are wide fan beams that branch off to blanket user areas with coverage from multiple angles.
In the work presented at SIGCOMM 2026, the team proves mathematically that this two-tier structure emerges naturally from the physics of passive relaying, and is near-optimal for maximizing both coverage and robustness.
Once fabricated and mounted on walls or ceilings, the tiles with metasurfaces behave like natural reflectors to standard mmWave radios. No firmware changes, no modifications to standard network protocols, no awareness of the metasurface network required.
The engineers highlight that their networks of fixed metasurfaces still serve mobile users. Access points already scan across their steerable beams many times a second, and FlowForm designs the metasurfaces so that every user location is reachable through several surfaces at different angles. As a person walks or turns, the access point simply lands on whichever subset of surfaces currently delivers the highest rate, without knowing the surfaces are there. The same extra paths also give standard protocols more opportunities to schedule concurrent transmissions to multiple users without interference.
The team validated FlowForm across five real indoor environments using a mmWave radio testbed. FlowForm nearly doubles average link rates and more than doubles coverage area in challenging environments, matching idealized active RIS performance at a fraction of the cost of conventional fixes to the millimeter wave path-obstruction problem. These fixes — including denser access points, active relays, and reconfigurable intelligent surfaces (RIS) — are expensive. They require power supplies, control channels, and coordination protocols, which are cost prohibitive and hold back scaling. A single active mmWave RIS device, for example, can easily exceed thousands of dollars.
Zhang notes that the UC San Diego team has a provisional patent for technology and that they are open to collaborating with industry partners for technology licensing or future commercialization pathways.
Paper Information
The paper " FlowForm: Scalable Passive Metasurface Network for mmWave Coverage Expansion "will be presented at ACM SIGCOMM '26, on August 19 in Denver, CO.
Funding for the project came from the OUSD (R&E)/RT&L, the Office of the Under Secretary of Defense for Research and Engineering, which is part of the US Department of Defense/War.
Learn more at the project page: flowform.wqzhao.org .
