Rotational motion-aware beam refinement for high-throughput mmWave communications

Tourangbam Harishore Singh, Shabirahmed Badashasab Jigalur, Po Tsang Huang*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Millimeter-wave (mmWave) communication is used in high-throughput wireless networks. The shorter wavelength of mmWave requires the use of highly directional beamforming transmission techniques. However, transmission link degradation is significant with fast movement in the end user (EU) device. In this study, a rotational motion-aware beam refinement procedure (BRP) was proposed to maintain high-throughput data transmission during data transmission intervals (DTIs). Within the millisecond time frame of a DTI, the vulnerable rotational movement of the EU device can be detected using a three-axis orientation sensor (gyro). Therefore, the proposed BRP algorithm was employed at the EU device to switch the beam and thus achieve a proper line-of-sight. The process of beam switching is performed by directly detecting changes in the orientation of the EU device. The experimental results demonstrated that the proposed algorithm maintains high-throughput transmission under the fast movement of a EU device. The execution time of the BRP algorithm can be as low as 10 µs. Therefore, the overhead is solely the result of the sampling frequency of the motion sensors. The simulation results demonstrated that the use of our proposed algorithm improved the average received signal strength indicator by over 10 dBm. Furthermore, an average throughput of 928 Mbps is achieved with hardware testing, which is close to the maximum throughput of 970 Mbps for a low-power 60-GHz mmWave-embedded device.

Original languageEnglish
JournalWireless Networks
DOIs
StateAccepted/In press - 2021

Keywords

  • Beam refinement
  • mmWave
  • Motion-aware

Fingerprint Dive into the research topics of 'Rotational motion-aware beam refinement for high-throughput mmWave communications'. Together they form a unique fingerprint.

Cite this