Research On Reliability Technology In New Generation Mobile Wireless Network

In 5G networks,prediction of service reliability is critical because of strict service performance requirements.By combining artificial intelligence and machine learning,next-generation cellular systems will enable advanced data analysis technologies for efficient service quality management and network automation.Based on this idea,this thesis proposes a wireless service model with reliability calculation,which uses a Bayesian Network(BN)to compactly represent joint probability distribution.This thesis first describes the modeling of wireless networks to generate input data for data acquisition and machine learning algorithms,and then describes how to apply Bayesian network representation to provide a feasible calculation process,thus solving this prediction problem.In the Bayesian network learning phase studied,the characteristics of the base station's load,user position and moving speed affect the user's received signal-to-noise ratio(SNR)and signal interference +noise ratio(SINR).The test result is the probability or number of times the mobile user's throughput meets the threshold.Computer simulation results show that the model can predict network service reliability and infer the hidden state of the network under low-speed moving conditions.Aiming at the influence of the shadow effect in the wireless channel on the quality of the end user link in the cell,this thesis proposes a Bayesian network with hidden nodes to solve the reliability prediction scheme.In the process of end user movement,wireless systems usually perform real-time data collection.In the shaded area,the received signal-to-noise ratio may be lower than a predetermined threshold or cause a signal interruption.At this time,the data acquired by the acquisition system will become incomplete.Therefore,parameter learning of the network is performed by using an EM(Expectation Maximization)algorithm.The computer simulation results show that for the shadow region on the moving track,the Bayesian network service model with hidden nodes can well predict and infer the hidden state of the network,thus achieving efficient network reliability service and automation.