Linearization Trade-Offs in a 5G mmWave Active Phased Array OTA Setup

Feridoon Jalili*, Felice Francesco Tafuri, Ole Kiel Jensen, Yunfeng Li, Ming Shen, Gert Frølund Pedersen

*Corresponding author

Research output: Contribution to journalJournal articleResearchpeer-review

31 Downloads (Pure)

Abstract

The new generation of 5G mobile communication systems is using millimeter wave (mmWave) active phased arrays (APA) which have up to hundreds of individual analog transmitter and receiver chains and antennas. For these highly integrated systems linearization of each analog path is very challenging. Therefore a single input single output (SISO) system in combination with over the air (OTA) measurement is considered as an efficient approach for linearization. However, the knowledge about the dependency of the total SISO nonlinearity on the contributions from different blocks in the antenna array, as well as the linearization trade-offs is still missing. In this paper, an overview of the possible linearization trade-offs in an OTA setup with a mmWave APA is provided. The linearization technique is applied to a 4times 4 active phased array containing up-conversion of a sub 6 GHz LTE10 signal to an RF frequency of 28 GHz. Through measurements, the effects on adjacent channel power ratio (ACPR) and error vector magnitude (EVM) have been investigated for the following scenarios: i. impact from the up-converter, ii. impact of the steering angle due to antenna crosstalk and iii. a linearity comparison between a linearized and a backed-off system.

Original languageEnglish
Article number9116970
JournalIEEE Access
Volume8
Pages (from-to)110669-110677
Number of pages9
ISSN2169-3536
DOIs
Publication statusPublished - 18 Jun 2020

Keywords

  • Active phased array (APA)
  • digital pre-distortion (DPD)
  • millimeter wave (mmWave)
  • over the air (OTA)
  • power amplifier (PA)
  • single input single output (SISO)

Fingerprint Dive into the research topics of 'Linearization Trade-Offs in a 5G mmWave Active Phased Array OTA Setup'. Together they form a unique fingerprint.

Cite this