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

Download:

Current browse context:

eess

Change to browse by:

References & Citations

Bookmark

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

Computer Science > Information Theory

Title: An Efficient Code-Domain NOMA Transceiver for mm-Wave Hybrid Massive MIMO Architectures

Abstract: Massive MIMO and non-orthogonal multiple access (NOMA) are crucial methods for future wireless systems as they provide many advantages over conventional systems. Power-domain NOMA methods are investigated in massive MIMO systems, whereas there is little work on integration of code-domain NOMA and massive MIMO which is the subject of this study. We propose a general framework employing user-grouping based hybrid beamforming architecture for mm-wave massive MIMO systems where NOMA is considered as an intra-group process. It is shown that classical receivers of sparse code multiple access (SCMA) and multi-user shared access (MUSA) can be directly adapted. Additionally, a novel receiver architecture which is an improvement over classical one is proposed for uplink MUSA. This receiver makes MUSA preferable over SCMA for uplink transmission with lower complexity. We provide a lower bound on achievable information rate (AIR) as a performance measure. We show that code-domain NOMA schemes outperform conventional methods with very limited number of radio frequency (RF) chains where users are spatially close to each other. Furthermore, we provide an analysis in terms of bit-error rate and AIR under different code length and overloading scenarios for uplink transmission where flexible structure of MUSA is exploited.
Comments: Partially presented at IEEE ICC 2020 Workshop on NOMA for 5G and Beyond and to be submitted to IEEE Transactions on Communications, v2: Title changed, typos corrected and references added
Subjects: Information Theory (cs.IT); Signal Processing (eess.SP)
Cite as: arXiv:2007.04056 [cs.IT]
  (or arXiv:2007.04056v2 [cs.IT] for this version)

Submission history

From: Murat Bayraktar [view email]
[v1] Wed, 8 Jul 2020 12:10:30 GMT (811kb,D)
[v2] Mon, 12 Oct 2020 09:46:07 GMT (713kb,D)

Link back to: arXiv, form interface, contact.