2021
DOI: 10.1109/jsac.2021.3071836
|View full text |Cite
|
Sign up to set email alerts
|

Multi-Hop RIS-Empowered Terahertz Communications: A DRL-Based Hybrid Beamforming Design

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
150
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
5
3
2

Relationship

1
9

Authors

Journals

citations
Cited by 336 publications
(150 citation statements)
references
References 56 publications
0
150
0
Order By: Relevance
“…In this method, by observing the predefined rewards with state and action spaces, the joint design of active and passive beamforming is optimized through trial-and-error interactions with the environment. Specifically, DRL based methods were applied in the RIS-aided multi-user systems to solve spectrum-efficiency optimization problem in [98], secure capacity optimization problem in [99], energy-efficiency optimization problem in [100], and the Terahertz communication in [101].…”
Section: Deep Reinforcement Learning Based Methodsmentioning
confidence: 99%
“…In this method, by observing the predefined rewards with state and action spaces, the joint design of active and passive beamforming is optimized through trial-and-error interactions with the environment. Specifically, DRL based methods were applied in the RIS-aided multi-user systems to solve spectrum-efficiency optimization problem in [98], secure capacity optimization problem in [99], energy-efficiency optimization problem in [100], and the Terahertz communication in [101].…”
Section: Deep Reinforcement Learning Based Methodsmentioning
confidence: 99%
“…The system model of the DS-FTTD architecture at the m th carrier can be expressed as We consider that the CSI has been obtained, while the impact of imperfect CSI is numerically evaluated in Sec. V. In this work, we focus on the hybrid beamforming at transmitter, while one potential future extension is the use of reconfigurable intelligent surfaces (RIS) to manipulate the communication environment, which not only improves the coverage distance [46] but also reduces the complexity of beamforming at the transmitter [47].…”
Section: A Problem Formulation For Ds-fttd Hybrid Beamformingmentioning
confidence: 99%
“…III-C) in a multi-hop network [248]. For example, simulation results illustrate the feasibility and advancement of multi-hop THz RIS assisted networks, compared to single-hop systems and those without RIS [249]. Specifically, the management of direction is supported by deep reinforcement learning (DRL) to achieve the optimization for NP-hard hybrid beamforming.…”
Section: Multi-hop Communications With Active and Passive Relaysmentioning
confidence: 99%