An optimal multitier resource allocation of cloud RAN in 5G using machine learning

The networks are evolving drastically since last few years in order to meetuser requirements. For example, the 5G is offering most of the available spec-trum under one umbrella. In this work, we will address the resource allocationproblem in fifth-generation (5G) networks, to be exact in the Cloud Radio AccessNetworks (C-RANs). The radio access network mechanisms involve multiplenetwork topologies that are isolated based on the spectrum bands and it shouldbe enhanced with numerous access technology in the deployment of 5G net-work. The C-RAN is one of the optimal technique to combine all the availablespectral bands. However, existing C-RAN mechanisms lacks the intelligence per-spective on choosing the spectral bands. Thus, C-RAN mechanism requires anadvanced tool to identify network topology to allocate the network resources forsubstantial traffic volumes. Therefore, there is a need to propose a frameworkthat handles spectral resources based on user requirements and network behav-ior. In this work, we introduced a new C-RAN architecture modified as multitierHeterogeneous Cloud Radio Access Networks in a 5G environment. This archi-tecture handles spectral resources efficiently. Based on the simulation analysis,the proposed multitier H-CRAN architecture with improved control unit innetwork management perspective enables augmented granularity, end-to-endoptimization, and guaranteed quality of service by 15 percentages over theexisting system.