|
For Full-Text PDF, please login, if you are a member of IEICE,
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
|
A New User Selection Measure in Block Diagonalization Algorithm for Multiuser MIMO Systems
Riichi KUDO Yasushi TAKATORI Kentaro NISHIMORI Atsushi OHTA Shuji KUBOTA Masato MIZOGUCHI
Publication
IEICE TRANSACTIONS on Communications
Vol.E92-B
No.10
pp.3206-3218 Publication Date: 2009/10/01 Online ISSN: 1745-1345
DOI: 10.1587/transcom.E92.B.3206 Print ISSN: 0916-8516 Type of Manuscript: PAPER Category: Wireless Communication Technologies Keyword: multiuser MIMO, user selection, block diagonalization, channel capacity,
Full Text: FreePDF
Summary:
Multiuser -- Multiple Input Multiple Output (MU-MIMO) techniques were proposed to increase spectrum efficiency; a key assumption was that the Mobile Terminals (MTs) were simple with only a few antennas. This paper focuses on the Block Diagonalization algorithm (BD) based on the equal power allocation strategy as a practical MU-MIMO technique. When there are many MTs inside the service area of the access point (AP), the AP must determine, at each time slot, the subset of the MTs to be spatially multiplexed. Since the transmission performance depends on the subsets of MTs, the user selection method needs to use the Channel State Information (CSI) obtained in the physical layer to maximize the Achievable Transmission Rate (ATR). In this paper, we clarify the relationship between ATR with SU-MIMO and that with MU-MIMO in a high eigenvalue channel. Based on the derived relationship, we propose a new measure for user selection. The new measure, the eigenvalue decay factor, represents the degradation of the eigenvalues in null space compared to those in SU-MIMO; it is obtained from the signal space vectors of the MTs. A user selection method based on the proposed measure identifies the combination of MTs that yields the highest ATR; our approach also reduces the computational load of user selection. We evaluate the effectiveness of user selection with the new measure using numerical formulations and computer simulations.
|
|
|