Cognitive Neural Network Delay Predictor for High Speed Mobility in 5G C-RAN Cellular Networks

Abstract

The future fifth Generation (5G) cellular network is expected to support one million connections per square kilometre with 1 ms end-to-end as a desired latency. The potential of this ultra-dense network motivated the researchers to develop a new architecture. Cloud Radio Access Network (C-RAN) technology was proposed to meet the demand of future networks, however, moving the baseband processing from multiple physical base stations on the ground within the cell site into the cloud brings many challenges. One of these challenges is how to acquire accurate Channel State Information (CSI) for a dense number of access points and User Equipment (UE), which are the future theme of 5G deployment. CSI reflects the instantaneous communication link status between the mobile user and the base station. Hence, the imperfect or delayed CSI can influence the performance of the whole network. In order to reduce the impact of this outdated CSI and to improve its accuracy in C-RAN architecture, a Cognitive Neural Network Delay Predictor (CNNDP) is proposed for compensating the transmission and acquisition delay of the CSI working simultaneously along with the conventional prediction technique for predicting the time variations of the communication channel. The results demonstrate a significant enhancement in the data throughput of the network with the proposed approach.