Optimization of Resource Control Strategies for Heterogeneous UAV Elastic Optical Networks Under SDN Architecture

Abstract

With the widespread application of unmanned cluster technology, broadband, low latency, high flexibility, and reliable fifth-generation (5G) UAV communication networks are increasingly becoming a key issue. The traditional network technology faces several challenges, including irregular distribution of spectrum resources, real-time changes of network topology, and a variety of unforeseen services. The elastic optical networks (EONs) is integrated into UAV networks to effectively address resource fragmentation and optimize spectrum allocation with its high bandwidth, low latency and dynamic tunability in this article. A heterogeneous UAV-EON system is proposed to realize the collaborative management of network access and resource allocation. To obtain the approximate optimal solution of network access and routing and spectrum allocation (RSA) in UAV-EON system, this article presents a hybrid two-stage optimized algorithm which combines the global search function of whale optimization algorithm (WOA) with the local optimization function of genetic algorithm (GA), ensuring the consistency and effectiveness of the UAV network. Therefore, the algorithm can address the challenges of network selection, routing, and spectrum allocation in UAV-EON. Finally, we test the number of successful assignments and resource utilization under different workloads and network scale. Considering the high dynamic characteristics of UAV nodes and the fading characteristics of atmospheric laser channels, a real network scenario is constructed and a cross-layer optimal algorithm from physical layer to network is proposed. The research results show that compared with the traditional intelligent optimization algorithm, the proposed algorithm can improve by more than 10% in the success rate of task allocation and resource occupation.