TY - JOUR
T1 - Deep Reinforcement Learning based Joint Spectrum Allocation and Configuration Design for STAR-RIS-Assisted V2X Communications
AU - Aung , Pyae Sone
AU - Nguyen, Loc X.
AU - Tun, Yan Kyaw
AU - Han, Zhu
AU - Hong, Choong Seon
PY - 2024/4/1
Y1 - 2024/4/1
N2 - Vehicle-to-everything (V2X) communications is pivotal for modern transportation systems, but the challenges arise in scenarios with buildings, leading to signal obstruction and limited coverage. To alleviate these challenges, reconfigurable intelligent surface (RIS) is regarded as an effective solution for communication performance by tuning passive signal reflection. RIS has acquired prominence in 6G networks due to its improved spectral efficiency, simple deployment, and cost-effectiveness. Nevertheless, conventional RIS solutions have coverage limitations. Researchers are exploring on the promising concept of simultaneously transmitting and reflecting RIS (STAR-RIS), which provides 360° coverage while utilizing the advantages of RIS technology. In this article, an STAR-RIS-assisted V2X communication system is investigated. An optimization problem is formulated to maximize the achievable data rate for vehicle-to-infrastructure (V2I) users while satisfying the latency and reliability requirements of vehicle-to-vehicle (V2V) pairs by jointly optimizing the spectrum allocation, amplitude and phase shift values of STAR-RIS elements, digital beamforming vectors for V2I links, and transmit power for V2V pairs. Since it is challenging to solve in polynomial time, we decompose our problem into two subproblems. For the first subproblem, we model the control variables as a Markov Decision Process and propose a combined double deep Q-network (DDQN) with an attention mechanism so that the model can potentially focus on relevant inputs. For the latter, a standard optimization-based approach is implemented to provide a real-time solution, reducing computational costs. Numerical results demonstrate that our solution approach outperforms the vanilla DDQN approach by 5.2%, and our proposed system outperforms the conventional RIS by 39%.
AB - Vehicle-to-everything (V2X) communications is pivotal for modern transportation systems, but the challenges arise in scenarios with buildings, leading to signal obstruction and limited coverage. To alleviate these challenges, reconfigurable intelligent surface (RIS) is regarded as an effective solution for communication performance by tuning passive signal reflection. RIS has acquired prominence in 6G networks due to its improved spectral efficiency, simple deployment, and cost-effectiveness. Nevertheless, conventional RIS solutions have coverage limitations. Researchers are exploring on the promising concept of simultaneously transmitting and reflecting RIS (STAR-RIS), which provides 360° coverage while utilizing the advantages of RIS technology. In this article, an STAR-RIS-assisted V2X communication system is investigated. An optimization problem is formulated to maximize the achievable data rate for vehicle-to-infrastructure (V2I) users while satisfying the latency and reliability requirements of vehicle-to-vehicle (V2V) pairs by jointly optimizing the spectrum allocation, amplitude and phase shift values of STAR-RIS elements, digital beamforming vectors for V2I links, and transmit power for V2V pairs. Since it is challenging to solve in polynomial time, we decompose our problem into two subproblems. For the first subproblem, we model the control variables as a Markov Decision Process and propose a combined double deep Q-network (DDQN) with an attention mechanism so that the model can potentially focus on relevant inputs. For the latter, a standard optimization-based approach is implemented to provide a real-time solution, reducing computational costs. Numerical results demonstrate that our solution approach outperforms the vanilla DDQN approach by 5.2%, and our proposed system outperforms the conventional RIS by 39%.
KW - Array signal processing
KW - Optimization
KW - Reinforcement learning
KW - Resource management
KW - Spectral efficiency
KW - V2X communication
KW - Vehicle-to-everything
KW - Wireless communication
KW - attention mechanism
KW - deep reinforcement learning (DRL)
KW - double deep q-network (DDQN)
KW - reconfigurable intelligent surface (RIS)
KW - simultaneously transmitting and reflecting RIS (STAR-RIS)
UR - http://www.scopus.com/inward/record.url?scp=85180344187&partnerID=8YFLogxK
U2 - 10.1109/JIOT.2023.3329893
DO - 10.1109/JIOT.2023.3329893
M3 - Journal article
SN - 2327-4662
VL - 11
SP - 11298
EP - 11311
JO - IEEE Internet of Things Journal
JF - IEEE Internet of Things Journal
IS - 7
ER -