Reseach On A Fine-grained Cloud-RAN Framework And Key Technique

Cloud-RAN architecture as a new concept of green wireless communication systems, once be presented attracted the attention of many researchers and manufacturers. Cloud-RAN architecture’s main idea is to separate the RF unit and baseband processing unit. Then locate the baseband processing unit centrally. It becomes a virtual “cloud base”. The baseband processing unit having such a structure has the ability to share resources. Currently, research on Cloud-RAN architecture was based on the entire baseband link as the smallest scheduling unit. And these researches didn’t in-depth analysis the possible parallelism in different signal algorithm modules. This article dedicated to research the fine-grained Cloud-RAN architecture and the scheduling algorithms.Firstly, this article describes the current researches on Cloud-RAN architecture and proposed a fine-grained Cloud-RAN architecture. This architecture schedules the signal processing algorithm as the minimal scheduling unit in wireless communication system. Then this article analysis the various factors affecting the data processing delay in this architecture and propose the data delay consideration equation. According to the wireless communication system delay constraint,Furthermore, we make some reasonable assumptions for signal processing module and use the classical queuing theory and network queuing theory to mathematical model the fine-grained Cloud-RAN to get the performance indicators of the Cloud-RAN system. Then solve the waiting time probability distribution function and the average function of waiting time to get the signal processing module scheduling solution.Finally, in this article simulate the system latency using our proposed scheduling algorithm in different conditions and establish the uplink and downlink of OFDM modulation system on general purpose processor platform proved the feasibility of the system.