Reconfigurable Intelligent Surface Enabled Vehicular Communication: Joint User Scheduling and Passive Beamforming

Abstract

Given its ability to control and manipulate wireless environments, reconfigurable intelligent surface (RIS), also known as intelligent reflecting surface (IRS), has emerged as a key enabler technology for the six-generation (6G) cellular networks. In the meantime, vehicular environment radio propagation is negatively influenced by a large set of objects that cause transmission distortion such as high buildings. Therefore, this work is devoted to explore the area of RIS technology integration with vehicular communications while considering the dynamic nature of such communication environment. Specifically, we provide a system model where RoadSide Unit (RSU) leverages RIS to provide indirect wireless transmissions to disconnected areas, known as dark zones. Dark zones are spots within RSU coverage where the communication links are blocked due to the existence of blockages. In details, a discrete RIS is utilized to provide communication links between the RSU and the vehicles passing through out-of-service zones. Therefore, the joint problem of RSU resource scheduling and RIS passive beamforming or phase-shift matrix is formulated as an optimization problem with the objective of maximizing the minimum average bit rate. The formulated problem is mixed integer non-convex program which is difficult to be solved and does not account for the uncertain dynamic environment in vehicular networks. Thereby, we resort to alternative methods based on Deep Reinforcement Learning to determine RSU wireless scheduling and Block Coordinate Descent (BCD) to solve for the phase-shift matrix, i.e., passive beamforming, of the RIS. The Markov Decision Process (MDP) is defined and the complexity of the solution approach is discussed. Our numerical results demonstrate the superiority of our proposed approach over baseline techniques.

Machine Learning Approaches for Reconfigurable Intelligent Surfaces: A Survey

“…To provide good quality of network service for traffic moving through an obstructed area, Al-Hilo et al [100] proposed a DRL framework with multi-binary action space to search a policy that maximizes the minimum average bit rate for vehicles using wireless scheduling. The authors suggested a route with no direct connectivity via a roadside unit (RSU).…”

Pervasive Machine Learning for Smart Radio Environments Enabled by Reconfigurable Intelligent Surfaces

“…Although MDP and POMDP formulations are the most popular for wireless communications up to date [103], [161], [162], [163], [164], [165], [166], [167], [168], [169], [170], [171], [172], [173], the multiarmed bandits' formulations have a substantial proposition of value, despite their simplicity. The multiarmed bandits' algorithms are more intuitive and explainable compared to DRL methods.…”

“…Although the DRL-based design formulation in this work is quite general to account for a large number of applications, it is worth mentioning that more elaborate systems do not adhere to this framework. Like [165], the works [171], [172], [173] considered mobile RISs and opted for traditional optimization algorithms for their orchestration, which differs from the general formulation presented herein. Specifically, Al-Hilo et al [171] dealt…”

AIRIS: Artificial Intelligence Enhanced Signal Processing in Reconfigurable Intelligent Surface Communications

“…Application scenarios Environmental sensing Supervised learning [15] Localization problem in the system with only one AP and one RIS Machine learning [16] Image-based sensing with a high resolution image of the propagation environment Channel acquisition Deep residual learning [17] Direct cascaded channel estimation over the RIS communications Federated learning [18] Channel estimation in multi-user case Supervised learning [19] Antenna domain cascaded channel extrapolation with the full passive RIS Supervised learning [22] Antenna domain individual channel extrapolation with the hybrid active/passive RIS Beamforming design Supervised learning [27] Beamforming design with passive RIS under the indoor scenario Deep reinforcement learning [28] Beamforming design with the learning of the environment Supervised learning [32] Solving the non-convex problem to maximize the system sum-rate in the RIS based hybrid precoding system Supervised learning [22] Simultaneously optimizing the RIS activation pattern and the RIS beamforming in the hybrid passive/active RIS system Supervised learning [34] Beam tracking under the high mobility scenario Resource scheduling Deep reinforcement learning [35] User partitioning and RIS beamforming for the RIS assisted NOMA networks Supervised learning [36] Improving the performance of RIS assisted mmWave communications Deep reinforcement learning [37] Joint scheduling of multiple RISs to enlarge the sensing range of the RIS assisted communications Deep reinforcement learning [38] Maximizing the average energy efficiency through joint optimizing the power allocation, RIS beamforming, and RIS elements' ON/OFF state Federated learning [39] Privacy-preserving design in the RIS assisted mmWave system Federated learning [40] Minimizing the aggregation error and accelerating the convergence rate under the static scenario with multiple RIS Multi-agent deep reinforcement learning [41] Dynamic control of the RISs under the mobility scenario Deep reinforcement learning [42] Joint terminal scheduling and passive beamforming over the RIS empowered vehicular communications end, we will introduce AI assisted environmental sensing, AI assisted channel acquisition, AI assisted beamforming design, and AI assisted resource scheduling in different scenarios, as shown in Table I. Except for the state-of-the-art researches on AIRIS, we will also present some future research directions aiming to provide diverse technical discussions and facilitate the continuous research on AIRIS.…”

“…If same traffic is offered to all the vehicles within the dark zones, the low mobility vehicles will achieve poor service quality than the high ones. Hence, in [42], the joint vehicle scheduling and passive beamforming in RIS empowered vehicular communication are investigated to maximize the minimum achievable rate for the vehicles in the dark zone. Specially, the whole problem is decoupled into two sub-ones: the wireless scheduling and the RIS phase-shift optimization.…”