Channel Performance Metrics and Evaluation for XR Head-Mounted Displays with mmWave Arrays
arXiv:2404.19297 · doi:10.1109/TCOMM.2024.3392805
Abstract
Millimeter-wave (mmWave) technology holds the potential to revolutionize head-mounted displays (HMDs) by enabling high-speed wireless communication with nearby processing nodes, where complex video rendering can take place. However, the sparse angular profile of mmWave channels, coupled with the narrow field of view (FoV) of patch-antenna arrays and frequent HMD rotation, can lead to poor performance. We introduce six channel performance metrics to evaluate the performance of an HMD equipped with mmWave arrays. We analyze the metrics using analytical models, discuss their impact for the application, and apply them to 28 GHz channel sounding data, collected in a conference room using eight HMD patch-antenna arrays, offset by 45 degrees from each other in azimuth. Our findings confirm that a single array performs poorly due to the narrow FoV, and featuring multiple arrays along the HMD's azimuth is required. Namely, the broader FoV stabilizes channel gain during HMD rotation, lessens the attenuation caused by line of sight (LoS) obstruction, and increases the channel's spatial multiplexing capability. In light of our findings, we conclude that it is imperative to either equip the HMD with multiple arrays or, as an alternative approach, incorporate macroscopic diversity by leveraging distributed access point (AP) infrastructure.
References in corpus (6)
- Indoor 5G 3GPP-like Channel Models for Office and Shopping Mall Environments
- Toward 6G with Terahertz Communications: Understanding the Propagation Channels
- Enhanced Effective Aperture Distribution Function for Characterizing Large-Scale Antenna Arrays
- Opportunities and Challenges for Virtual Reality Streaming over Millimeter-Wave: An Experimental Analysis
- Impact of Array Configuration on Head-Mounted Display Performance at mmWave Bands
- DOPAMINE: Doppler frequency and Angle of arrival MINimization of tracking Error for extended reality