【CCF-AIR青年基金】High Density and High Performance mmWave Networking System

Research Themes

Others

Background

Immersive virtual environments such as Metaverse and extended reality (xR) demand high-performance wireless networking in the order of 1Gbps throughput and 10ms latency. The challenge is further amplified when there are a dozen or even more such devices within one room. How do we build a high-density and high-performance wireless networking infrastructure to support such applications? According to the current development of 5G as well as the future trend of 6G, exploiting higher-frequency wireless techniques is one of the mainstream ideas to alleviate this problem. Among them, mmWave with 30GHz~300GHz carrier frequency may be the potential answer due to its broader available spectrum for a higher channel capacity and it narrower RF beam for more directional connections. These exquisite features are not conveniently available in sub-6GHz systems due to the strict spectrum regulation and the large antenna size.

 

However, it's a non-trivial task to build a high-density and high-performance mmWave networking system with the state-of-the-art techniques. We believe the management of the mmWave resources becomes more challenging: despite conventional multiplexing methods in the time and frequency domain, there is an additional degree of freedom - space multiplexing - as the result of directional beams in mmWave systems. The additional DoF is a double-edged sword, which could on one side enable denser links when properly used, but on the other side create headaches of discovering, assigning, and tracking beams in both static and dynamic environment. Moreover, the careful design of interference management among multiple access points and end-user devices is also a necessity. All these have to be done in an efficient manner to ensure that the disadvantages do not cancel out all the advantages.

 

Despite confining the management on the PHY layer of each device, the design of higher layers is also indispensable to optimize the system performance. For instance, we believe that a centralized higher-layer design that is tailored to each application scenario would enable sharing information and control in a cross-layer, cross-device manner, thus making it possible for higher overall performance and directly optimized for target applications. As a starting point, one or more sample applications can be chosen to make it straightforward to design and benchmark the performance of the wireless network and the application experience. Sample application scenarios include Metaverse in an office environment or xR gaming in a store setting.

Target

  • A large-scale mmWave testbed consists of up to hundreds of devices and multiple access points with full access to low-level PHY control on both devices.
  • A management system enables centralized application/scenario-aware orchestration of mmWave resources, achieving efficient, low interference scheduling of mmWave resources.
  • Throughout tests with demo applications showing tens to hundreds of Gbps aggregated throughput in a room-scale while meeting stringent application requirements such as xR and metaverse.

Related Research Topics

  • Indoor wireless environment sensing and modeling.
  • Beam generating and steering for the phased-array device or the distributed phased arrays.
  • Multipath effect exploitation for throughput enhancement.
  • Distributed wireless networking resource management.
  • Large-scale networking testbed construction.
  • Future networking infrastructure for xR and Metaverse.

Scan QR code
关注Ali TechnologyWechat Account