Virtual Network Embedding Based on Multidimensional Resource Constraints in Heterogeneous Space-Air-Ground Network

Abstract

Network virtualization technology is one of the most effective methods to solve the problem of network ossification in future network, which allows for heterogeneous virtual network (VN) users to share the underlying network resources without changing the physical network infrastructures, and provides higher flexibility and better maintainability. Existing works mainly consider single-network resources such as a CPU (central processing unit) or bandwidth in virtual network embedding but ignore the problem of resource diversity in heterogeneous space-air-ground networks. Aiming at the problem of multidimensional resource constraints in virtual network embedding, two heuristic VN embedding algorithms are proposed in this passage. First, a novel VN two-stage algorithm (TE-VNE) according to the Topology attributes and multidimensional resource Eigenvalue attributes of heterogeneous networks such as storage, sensor, and frequency. Second, we propose a node resource similarity virtual network embedding (NRS-VNE). This is based on the similarity of resource eigenvalues to balance the competition between the main resources that may occur in the node mapping stage. Conclusion the experiments show that our algorithms work better in embedding acceptance and long-term average revenue of infrastructure Providers (InPs) and Service Provides (SPs) than single-resource attributes.