Deployment Strategies for Large Intelligent Surfaces
(2022) In IEEE Access 10. p.61753-61768- Abstract
Beyond 5G communication systems must be able to meet the requirements imposed by the ever-increasing demand in capacity, while guaranteeing robustness, reliability, low latency, security, as well as spectral and power efficiencies. Large intelligent surfaces (LIS) as an evolution of massive MIMO have drawn considerable attention among researchers, being already considered as one of the key technologies to be included in beyond 5G communication systems. Due to the massive number of antennas, it also brings several challenges namely in terms of computational complexity. In this paper, we intend to provide guidelines for the LIS practical implementation and configuration by specifying system parameters and their consequent relationship for... (More)
Beyond 5G communication systems must be able to meet the requirements imposed by the ever-increasing demand in capacity, while guaranteeing robustness, reliability, low latency, security, as well as spectral and power efficiencies. Large intelligent surfaces (LIS) as an evolution of massive MIMO have drawn considerable attention among researchers, being already considered as one of the key technologies to be included in beyond 5G communication systems. Due to the massive number of antennas, it also brings several challenges namely in terms of computational complexity. In this paper, we intend to provide guidelines for the LIS practical implementation and configuration by specifying system parameters and their consequent relationship for a panel-based LIS. In particular, the interplay between the number of baseband outputs per square metre, the fraction of activated area, the panel size and terminal density is summarised by an empirical law under the assumption that all terminals experience reasonable quality of service. Furthermore, performance results show that, in general, moderate panel sizes offer the best rates, highlighting that there is no need to activate a large fraction of LIS to provide an acceptable minimum terminal rate. However, such fractions may require more baseband outputs per panel, leading to a higher number of baseband outputs per square metre, translating into higher implementation complexity. Finally, it is observed that the implicit rate loss of using sparse static panel deployments instead of contiguous panel deployments that are dynamically activated/deactivated is not so significant, omitting the complexity involved in managing the set of activated panels.
(Less)
- author
- Pereira, Andreia ; Rusek, Fredrik LU ; Gomes, Marco and Dinis, Rui
- organization
- publishing date
- 2022
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- beyond 5G systems, dynamic resource allocation, Large intelligent surfaces (LIS), massive MIMO, spacial resource allocation
- in
- IEEE Access
- volume
- 10
- pages
- 16 pages
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- external identifiers
-
- scopus:85132681880
- ISSN
- 2169-3536
- DOI
- 10.1109/ACCESS.2022.3181757
- language
- English
- LU publication?
- yes
- id
- 42f15ed4-9131-4057-a8ba-5e825c130b49
- date added to LUP
- 2022-09-06 15:14:55
- date last changed
- 2023-11-21 11:05:46
@article{42f15ed4-9131-4057-a8ba-5e825c130b49,
abstract = {{<p>Beyond 5G communication systems must be able to meet the requirements imposed by the ever-increasing demand in capacity, while guaranteeing robustness, reliability, low latency, security, as well as spectral and power efficiencies. Large intelligent surfaces (LIS) as an evolution of massive MIMO have drawn considerable attention among researchers, being already considered as one of the key technologies to be included in beyond 5G communication systems. Due to the massive number of antennas, it also brings several challenges namely in terms of computational complexity. In this paper, we intend to provide guidelines for the LIS practical implementation and configuration by specifying system parameters and their consequent relationship for a panel-based LIS. In particular, the interplay between the number of baseband outputs per square metre, the fraction of activated area, the panel size and terminal density is summarised by an empirical law under the assumption that all terminals experience reasonable quality of service. Furthermore, performance results show that, in general, moderate panel sizes offer the best rates, highlighting that there is no need to activate a large fraction of LIS to provide an acceptable minimum terminal rate. However, such fractions may require more baseband outputs per panel, leading to a higher number of baseband outputs per square metre, translating into higher implementation complexity. Finally, it is observed that the implicit rate loss of using sparse static panel deployments instead of contiguous panel deployments that are dynamically activated/deactivated is not so significant, omitting the complexity involved in managing the set of activated panels.</p>}},
author = {{Pereira, Andreia and Rusek, Fredrik and Gomes, Marco and Dinis, Rui}},
issn = {{2169-3536}},
keywords = {{beyond 5G systems; dynamic resource allocation; Large intelligent surfaces (LIS); massive MIMO; spacial resource allocation}},
language = {{eng}},
pages = {{61753--61768}},
publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
series = {{IEEE Access}},
title = {{Deployment Strategies for Large Intelligent Surfaces}},
url = {{http://dx.doi.org/10.1109/ACCESS.2022.3181757}},
doi = {{10.1109/ACCESS.2022.3181757}},
volume = {{10}},
year = {{2022}},
}