PSO-Based Dynamic RSU Role Assignment Framework for Scalable and Reliable Content Delivery in VANETs

基于粒子群优化算法的动态路侧单元角色分配框架,用于在车载自组织网络中实现可扩展和可靠的内容交付

阅读:1

Abstract

Vehicular Ad-hoc Networks (VANETs) must sustain heterogeneous real-time content services, yet static roadside-unit (RSU) roles lead to congestion, coverage voids, and inefficient content delivery under bursty, concurrent demand. To address this issue, we propose a PSO-Based dynamic RSU role assignment framework named PDRA that dynamically adapts roles, coverage, and replication of RSU to current network conditions. A telemetry-based suitability estimator aggregates location, link stability, resource availability, traffic load, and content sensitivity at each RSU and feeds a Particle Swarm Optimization routine that assigns RSUs to Leader/Helper/Inactive roles while enforcing spatial separation between Leaders. An adaptive sectoring mechanism then resizes each cluster RSU's communication scope-contracting under overload to protect local latency and expanding during slack to assist neighbors-thereby suppressing void areas and balancing service density. On top of the physical layer of RSUs, Leader RSUs cooperatively form a virtual Replication Layer that maintains global visibility of load and content locality to steer requests and replicate popular data near demand, reducing backhaul reliance. Finally, a load- and energy-aware reconfiguration policy orchestrates staged assist/offload, selective Helper activation, PSO-based Leader reassignment, and sleep scheduling for underutilized RSUs, preserving resilience and sustainability. NS-3 urban scenarios corroborate that PDRA improves packet delivery, lowers end-to-end delay, reduces backhaul traffic, and increases fairness over strong baselines. By jointly optimizing role assignment, coverage control, and replication, PDRA offers a scalable and robust solution for VANET content delivery under dynamic, multi-user conditions.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。