Handover Performance Enhancement Method For High-speed Rail Mobile Network Under 5G C/U Separation Architecture

At present,high-speed rail has become one of the main ways of transportation for people to travel.People will spend more and more time on high-speed rail,which will lead to increasing demand for data transmission and higher performance requirements for high-speed rail mobile networks.With the rapid development of mobile communication networks,high-speed rail mobile networks are transitioning from 4G to 5G.To meet the needs of high-speed mobility,the 5G C/U separation network architecture is proposed,which achieves better transmission performance by adopting data plane and control plane separation technology.Even so,during data plane transmission,users still need to the handover of small cells,that is,the handover from the current cell to the target cell to achieve stable data transmission.In the mobile handover,it is very easy to cause problems such as data transmission failure and communication interruption.Although the existing research work has proposed some algorithms to solve this problem,the existing algorithms still have problems such as high handover delay,high failure rate and low data rate.To solve the above problems,this paper carried out the following research work:(1)Aiming at the problems of high handover delay and high handover failure rate,the predictive handover method based on Transformer is proposed to select the best handover point,thereby reducing handover delay and handover failure rate.Under the 5G C/U separation mobile network architecture,the mobile handover only occurs between the current base station and the next neighboring base station.To avoid the ping-pong effect in the handover process,the traditional A3 handover algorithm performs handover when the difference between the signal received reference power(RSRP)at three consecutive measurement moments exceeds the set threshold.The proposed scheme takes the measured RSRP from the current base station as input,uses the Transformer model to predict the RSRP of the next base station,and uses the predicted RSRP data to perform handover in advance,which can save the overhead of three measurement moments and reduce the handover delay.The experimental results show that compared with the traditional A3 handover algorithm,it can reduce the handover time by 56% and reduce the handover failure rate,thereby improving the network data transmission performance during the mobile handover process.(2)In the case that the handover point of the above research is known,and aiming at the problem of low user data rate immediately after handover,from the perspective of green network design,a resource optimization scheme assisted by time-domain power allocation and buffering is designed.Minimize system energy consumption under the premise of user requirements for network service quality.There are two types of users in the system: one is high real-time users,which have the lowest instantaneous data rate requirements,and the other is low real-time users,which have minimum data volume requirements within a certain period of time.Users with high real-time performance can ensure their needs through power allocation in the time domain,and users with low realtime performance can meet their needs by designing corresponding buffer mechanisms.Considering that the optimization goal is to minimize the total energy consumption while satisfying the user’s needs,an optimization problem is constructed under the constraints of the maximum transmit power of the base station,the minimum instantaneous data rate,the minimum cached data volume,and the release time of the cache.Since the problem is non-convex,an enumeration algorithm based on KKT(Karush-Kuhn-Tucker)conditions is proposed to solve it.The simulation verifies the effectiveness of the proposed algorithm and compares it with the time-domain power allocation water injection algorithm and the fixed power allocation algorithm.The results show that the time-domain power allocation water injection algorithm scheme cannot meet the needs of high real-time users,and the fixed power allocation algorithm consumes more energy.The proposed algorithm can meet the network service quality requirements of all users and obtain the lowest energy consumption.On the basis of the above research,aiming at maximizing energy efficiency,this paper designs a time-domain power allocation cache-assisted resource optimization scheme,and constructs the corresponding optimization problem.Since the problem is a fractional programming problem,the Dinkelbach-based solution algorithm is designed.The experimental results show that the proposed algorithm obtains the maximum energy efficiency value after 3 iterations and verifies the effectiveness of the proposed scheme.