Research On Base Station Location And Coverage Prediction Algorithm Based On MDT

In recent years,the development of mobile communication systems has undergone significant changes,posing severe challenges to the planning and optimization of wireless network.In order to have a more comprehensive grasp of the network situation and reduce operating costs,3GPP has studied and specified a solution in the R10 version,namely Minimization of Drive Tests(MDT).How to use MDT data to assist wireless network construction has become an important topic.In the work of wireless network planning and optimization,the longitude and latitude of base station is an important engineering parameter.At the same time,adjusting the antenna azimuth on the basis of accurate engineering parameters is a common means to improve the quality of network coverage.In view of the above situation,this paper will focuses on the base station location and coverage prediction algorithm based on MDT.Firstly,Introduce the key technologies and application scenarios of MDT,explain the principles of common wireless network positioning technology and the basic theory of propagation loss prediction to provide a theoretical basis for follow-up research.Secondly,based on the improvement of Taylor series expansion positioning algorithm,a base station positioning algorithm based on MDT is proposed.The Taylor positioning algorithm requires an initial iteration point.Choosing an initial point closer to the true value can ensure the iteration convergence and reduce the number of iterations.Therefore,this article adopts a positioning method based on Reference Signal Receiving Power(RSRP)to estimate the initial point.Since the reference path loss is unknown,the Weighted Least Square(WLS)formula is derived to jointly estimate the base station location and reference path loss.Since the reference path loss of each cell under the same base station is different,the dichotomy is used for each cell to solve the WLS problem.The average value of the estimated positions of all cells is taken as the initial iteration point.In addition,since the initial iteration point is relatively close to the true position of the base station,in order to further eliminate abnormal values,calculate the distances from User Equipments(UEs)to the initial point and the distances from the UEs to the base station respectively,and delete the UEs whose difference are too large.The results show that the base station positioning algorithm proposed in this paper is highly accurate and simple to implement,and can be used for actual base station position correction.Then,based on the MDT data before the antenna azimuth adjustment,a coverage prediction algorithm based on MDT is proposed to predict the RSRP after the azimuth adjustment.Based on the principle of constant spatial path loss,before and after the antenna azimuth adjustment,the RSRP changes only the part of the antenna gain.Therefore,for the same location,the RSRP before the antenna azimuth adjustment is used for point-to-point conversion to obtain the RSRP predicted value after the azimuth adjustment.For locations where there is no MDT data before the azimuth adjustment,the BP neural network is used to fit the RSRP before the adjustment.This method avoids the shortcomings of empirical models and semi-empirical models that are not high in accuracy in a small-scale range,and is simple to implement.In addition,this article is based on the reference model of the antenna directivity function in 3GPP TR25.996,and uses the least square method to fit the sampled data of the real pattern to obtain a more accurate directivity function.The test results show that the MDT-based coverage prediction algorithm proposed in this paper has high prediction accuracy,and the RSRP prediction value is closer to the measured data.Finally,Design and implement an online prediction system for mobile network coverage based on MDT.This paper uses WebGis technology to load MDT data,base station engineering parameters and electronic maps to the web page,and uses the coverage prediction algorithm based on MDT to predict the RSRP distribution after azimuth adjustment.The system is divided into five layers: the presentation layer is the web page,which is responsible for providing interactive operations for users;the control layer,the business logic layer and the data persistence layer are responsible for processing user requests and returning the results to the web page;the data layer is responsible for storing data,including MDT data,map resources,etc.The system was tested using MDT data,and the results showed that the system is simple to operate,and integrates the advantages of WebGis,such as lightweight,cross-platform,and low cost,which can provide help for the actual network planning and optimization work.