RAISE: Optimizing RIS Placement to Maximize Task Throughput in Multi-Server Vehicular Edge Computing

Yanan Ma, Zhengru Fang, Longzhi Yuan, Yiqin Deng · 2025

Read RAISE: Optimizing RIS Placement to Maximize Task Throughput in Multi-Server Vehicular Edge Computing by Yanan Ma, Zhengru Fang, Longzhi Yuan, Yiqin Deng (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-17708.txt.

Introduction

Self-driving cars must offload heavy computing to roadside servers, but trucks block signals and rush hours flood the network. Yanan Ma and co-authors propose RAISE, a framework that places reconfigurable intelligent surfaces, panels that steer radio waves, to open better paths. It tunes each panel's height and tilt together with task assignments across servers.

The aim is to push through as many tasks as possible before their deadlines. A two-layer method solves the problem: the inner layer assigns tasks optimally, and the outer layer climbs toward the best panel placement with low computing cost. Simulations beat existing benchmarks consistently. The authors present smart surface placement as a lever for vehicular edge computing capacity.

Catalog record

CREATORYanan Ma, Zhengru Fang, Longzhi Yuan, Yiqin Deng
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE68 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-17708

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

RAISE: Optimizing RIS Placement to Maximize Task Throughput in Multi-Server Vehicular Edge Computing Thanks: Y. Ma, Z. Fang, L. Yuan, Y. Deng, and Y. Fang are with Hong Kong JC Lab of Smart City and the Department of Computer Science, City University of Hong Kong, Hong Kong, China. E-mail: {yananma8-c, zhefang4-c}@my.cityu.edu.hk, {longyuan, yiqideng, my.fang}@cityu.edu.hk. Thanks: X. Chen is with the Department of Electrical and Electronic Engineering, University of Hong Kong, Hong Kong, China. E-mail: [email protected]. Yanan Ma, Zhengru Fang, Longzhi Yuan, Yiqin Deng Abstract Given the limited computing capabilities on autonomous vehicles, onboard processing of large volumes of latency-sensitive tasks presents significant challenges. While vehicular edge computing (VEC) has emerged as a solution, offloading data-intensive tasks to roadside servers or other vehicles is hindered by large obstacles like trucks/buses and the surge in service demands during rush hours. To address these challenges, Reconfigurable Intelligent Surface (RIS) can be leveraged to mitigate interference from ground signals and reach more edge servers by elevating RIS adaptively. To this end, we propose RAISE, an optimization framework for R IS pl a cement in mult i - s erver V E C systems. Specifically, RAISE optimizes RIS altitude and tilt angle together with the optimal task assignment to maximize task throughput under deadline constraints. To find a solution, a two-layer optimization approach is proposed, where the inner layer exploits the unimodularity of the task assignment problem to derive the efficient optimal strategy while the outer layer develops a near-optimal hill climbing (HC) algorithm for RIS placement with low complexity. Extensive experiments demonstrate that the proposed RAISE framework consistently outperforms existing benchmarks. I Introduction Connected and autonomous driving (CAD) is expected to profoundly revolutionize the future of transportation by delivering safer, more efficient, and more comfortable driving experiences [ 1 ] . To achieve this goal, autonomous vehicles (simply vehicles at later development) must gather huge amounts of real-time data from various onboard sensors and generate a vast number of computing tasks for smart driving and entertainment [ 2 ] , [ 3 ] , [ 4 ] . However, processing such massive multi-modal tasks within sub-seconds, as demanded by time-sensitive vehicular applications, presents a major hurdle that precludes the widespread deployment of CAD. Given the substantial deployment costs of computing power, a cost-effective solution is to offload some computing tasks from vehicles to multi-access edge computing (MEC) systems at roadside or more powerful vehicles located nearby [ 5 ] . With the assistance of MEC, the basic safety-critical tasks may still be processed onboard to ensure reliability and safety, whereas advanced driving tasks, such as trajectory planning, human-vehicle interactions, augmented reality (AR) navigation, and real-time interactive gaming, can be offloaded to MEC systems with more powerful computing/artificial intelligence (AI) capabilities. This strategy represents an emerging paradigm known as vehicular edge computing (VEC) [ 6 ] . VEC systems often handle data-intensive and delay-sensitive tasks, which require vehicles to establish fast communications with MEC servers that provide powerful computing capabilities [ 5 ] . Unfortunately, fulfilling these task requirements is fraught with significant challenges. On the one hand, the absence of line-of-sight (LoS) channels results in low data rates and significant transmission latency between vehicles and roadside VEC servers. The relatively low altitude of VEC servers, e.g., co-locating with base stations (BSs), access points (APs), and roadside units (RSUs), makes communication links vulnerable to physical blockages, including green belts, trucks, and buses. The situation is further exacerbated by the high mobility of vehicles, making vehicle-to-server connections highly unstable [ 7 ] . On the other hand, the high density of vehicles in populated areas, particularly during peak hours, causes a surge in data transmission and computation workload. The intensive communication-computation demands can easily overwhelm resource-limited roadside VEC servers, resulting in prolonged transmission and service waiting time. (a) Number of available servers versus RIS altitude. (b) Number of available servers versus RIS tilt angle. Fig. 1 : The impact of RIS placement on the average number of available servers for a vehicle. An available server refers to a server that can fulfill the offloaded tasks from the vehicle with a deadline requirement, which is set to 100 ms. In recent years, reconfigurable intelligent surface (RIS), an intelligent surface capable of augmenting transmission environments, has attracted significant attention [ 8 , 9 , 10 , 11 , 14 , 15 , 12 , 13 ] . Strategic deployment of RISs in VEC systems can effectively address the aforementioned challenges. First of all, elevated RISs, such as those mounted on lampposts or carried by unmanned aerial vehicles (UAVs), can establish relaying LoS links by reflecting signals from vehicles toward roadside servers [ 14 ] . Given the fact that direct links between vehicles and roadside servers are frequently obstructed in rapidly changing vehicular environments, these vehicle-RIS-server links can ensure high-speed and stable communications. Second, elevated RISs also facilitate vehicles to beam signals upwards, mitigating wireless interference caused by interlaced and congested electromagnetic (EM) signals close to the ground, thereby improving network-wide spectrum efficiency. Third, elevated RIS can help vehicles access more roadside servers by reflecting tasks to less loaded but potentially distant edge servers, thus achieving a better balance of communication and computation workloads in VEC systems [ 16 ] . Fourth, while elevated relays can serve a similar function, RISs merely reflect signals without introducing significant processing delays or half-duplex transmission delays [ 17 ] , making them more suitable for time-critical VEC applications. Finally, RIS deployment is cost-effective due to its lightweight design, ease of installation on walls or racks, and energy efficiency. In particular, RIS operates passively, requiring minimal energy for the phase shift controller and control signaling exchanges without the need to perform user data transmissions [ 15 ] . Based on the aforementioned observations, in this paper, we investigate the joint problem of RIS positioning and task offloading in multi-server VEC systems. As depicted in Fig. 1 , we consider a vehicle with a task that has a deadline requirement. The number of available servers (defined as the servers that can fulfill the task) initially increases with the RIS altitude, as a higher altitude provides a broader field of view. However, as altitude increases, communication link quality degrades due to path attenuation, resulting in prolonged communication delay. Thus, beyond a certain altitude, further increasing the altitude leads to a decline in the number of available servers that meet the overall task completion latency due to the prolonged communication delay. Besides, the RIS tilt angle requires careful adjustment because it should be precisely oriented toward the target areas to maximize connectivity. These observations emphasize the importance of carefully optimizing the RIS positioning, i.e., both the RIS altitude and tilt angle, to enhance the overall task computing performance and offer valuable insights into designing efficient and effective heuristic algorithms. Our objective in this paper is to optimize both the positioning of a RIS and task assignments to maximize task throughput , defined as the number of offloaded computing tasks that can be completed within their deadline constraints. A fundamental tradeoff in RIS placement involves balancing two conflicting factors of increasing the RIS altitude: a higher RIS enables vehicles to connect to more VEC servers by

Read the complete text