Channel Characterization for Wideband Large-Scale Antenna Systems Based on a Low-Complexity Maximum Likelihood Estimator

Research output: Contribution to journalJournal articleResearchpeer-review

24 Citations (Scopus)
430 Downloads (Pure)

Abstract

Wideband large-scale array systems operating at millimeter-wave bands are expected to play a key role in future communication systems. It is recommended by standardization groups to use sphericalwave models (SWMs) to characterize the channel in near-field cases because of the large array apertures and the small cell size. However, this feature is not widely reflected in channel models yet, mainly due to the high computational complexity of SWMs compared to that of the conventional plane-wave model (PWM), especially when ultrawideband signals are considered. In this paper, a maximum likelihood estimator (MLE) of low computational complexity is implemented with a SWM for ultrawideband signals. The measurement data obtained from an ultrawideband large-scale antenna array system at 28-30GHz are processed with the proposed algorithm. The power azimuth-delay profiles (PADP) estimated from the SWM and the PWM are compared to those obtained from rotational horn antenna measurement, respectively. It shows that the multipath components (MPCs) are well-estimated with the proposed algorithm, and significant improvement in estimation performance is achieved with the SWM compared to the PWM. Moreover, the physical interpretation of the estimated MPCs is also given along with the estimated scatterers.
Original languageEnglish
Article number8412215
JournalI E E E Transactions on Wireless Communications
Volume17
Issue number9
Pages (from-to)6018-6028
Number of pages11
ISSN1536-1276
DOIs
Publication statusPublished - 1 Sept 2018

Keywords

  • Spherical-wave signal model
  • channel estimation
  • millimeter wave
  • ultrawideband

Fingerprint

Dive into the research topics of 'Channel Characterization for Wideband Large-Scale Antenna Systems Based on a Low-Complexity Maximum Likelihood Estimator'. Together they form a unique fingerprint.

Cite this