We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Computer Science > Information Theory

Title: Bit-Interleaved Coded Multiple Beamforming in Millimeter-Wave Massive MIMO Systems

Abstract: An analysis of the diversity gain for bit-interleaved coded multiple beamforming (BICMB) method in millimeter-wave (mm-Wave) massive multiple-input multiple-output (MIMO) systems is carried out for both the single-user and multi-user scenario. We show that the diversity gain is independent of the number of data streams and full spatial multiplexing order can be achieved in both scenarios. Also, we show that the diversity gain in the multi-user scenario is independent of the number of users in the system and only depends on the number of the remote antenna units (RAUs) at the transmitter side, when each user has only one RAU. The assumption here is that the channel state information (CSI) is known at both sides of the transmitter and the receiver and the number of antennas in each RAU goes to infinity. This latter assumption can be relaxed by a large number of antennas in each RAU, similar to the case for all massive MIMO research. Based on these assumptions, the diversity gain for the single-user scenario is $\frac{\left(\sum_{i,j}\beta_{ij}\right)^2}{\sum_{i,j}\beta_{ij}^2L_{ij}^{-1}}$ where $L_{ij}$ is the number of propagation paths and $\beta_{ij}$ is the large scale fading coefficient between the $i$th RAU in the transmitter and the $j$th RAU in the receiver. The diversity gain in the multi-user scenario for the $k$-th user is $\frac{M^2}{\sum_{j}L_{kj}^{-1}}$ where $M$ represents the number of RAUs at the transmitter. Simulation results show that when the perfect channel state information assumption is satisfied, the use of BICMB results in the diversity gain values predicted by the analysis.
Comments: 26 pages, 11 figures. arXiv admin note: substantial text overlap with arXiv:1903.04693
Subjects: Information Theory (cs.IT)
Cite as: arXiv:1911.07297 [cs.IT]
  (or arXiv:1911.07297v1 [cs.IT] for this version)

Submission history

From: Sadjad Sedighi [view email]
[v1] Tue, 12 Nov 2019 23:38:52 GMT (531kb)

Link back to: arXiv, form interface, contact.