mmWall: steerable, transflective metasurface for mmWave networks

mmWave spectrum has emerged in the 5G era as a key next-generation wireless network enabler, fulfilling user demands for extremely high bandwidth networks. However, this technology faces significant headwinds. 5G outdoor coverage is difficult to bring indoors, as exterior building walls block mmWave signal. Currently, operators are forced to offload mmWave traffic onto lower frequencies or off their networks entirely (Wi-Fi) when users move indoors, incurring handover delays and application disruptions. Also, people, buildings, and other clutter block mmWave links, forcing data to flow over a much less reliable path for outdoor users. Moreover, mmWave has a narrow beam and works only when the transmitter’s beam is aligned with the receiver’s beam. Scanning the entire space to find the best alignment incurs a huge overhead. While mmWave networks sacrifice reliability for capacity, my work shows that they can coexist.

mmWall is the first reconfigurable intelligent surface that uses metamaterials to refract, reflect, and/or split incoming mmWave beams in desired directions. Its key technical contribution is enhancing reliability for mmWave networks. As a user enters the building, mmWall deployed on the window seamlessly refracts outdoor signals and precisely steers them toward the user indoors, making outdoor-to-indoor communication possible. When buildings occlude signals towards an outdoor user, mmWall provides a strong alternative path by reflecting signals towards the user. Furthermore, it splits the incoming beam and simultaneously sweep multi-armed beams to enable fast beam search. My extensive evaluation has shown that in this way, mmWall has the potential to dramatically improve reliability and spectral efficiency of networks as a whole.

Kun Woo Cho
Kun Woo Cho
Email: kwcho@princeton.edu

My research interests include wireless networked systems, smart surface, and AI-assisted networks.