A method for cell azimuth prediction based on mobile terminal data

Abstract

The invention relates to a method for realizing cell azimuth prediction based on mobile terminal data, which collects a wireless environment data set S1 with a location reported by an intelligent terminal within a period of time, and its records include important fields such as RSRP, TA, and AOA. Then, according to the statistical distribution of the AOA dimension data, calculate the central value AOAmid of AOA, and then extract the records whose AOA value is equal to AOAmid in the data set S1 according to the AOAmid to form the data set S2. Then filter out the non-direct signal records, for example, by establishing the actual distance model between TA and the sampling point to the location of the cell, and delete abnormal TA records according to the TA distance model to form the data set S3. Cluster the data set S3, cluster the data into K classes, take the class Ki with the strongest average signal RSRP, and the angle between the data center point of the class Ki and the north direction of the cell location is the predicted cell azimuth. This method reduces the operation cost, and its azimuth prediction accuracy is high.

Description

technical field [0001] The invention relates to the technical field of cell azimuth prediction, in particular to a method for realizing cell azimuth prediction based on mobile terminal data. Background technique [0002] The mobile communication network structure is a cellular implementation. In theory, a circle with the same radius covers the plane. When the center of the circle is in the center of each regular hexagon of the regular hexagonal grid, the number of circles used is the least. In a mobile communication network, the center of the circle is the location of the base station, and each base station is generally divided into three cells, and each cell has a fixed coverage azimuth and downtilt. [0003] During network planning, the azimuth and downtilt data of the cell are output through network simulation. After the station is built and in the later stage of station survey operation and maintenance, due to the need for coverage, the azimuth angle of the cell will ch...

AI Technical Summary

Problems solved by technology

Due to the existence of human factors, the accuracy of cell azimuth angle data in the current asset management system is not high. For example, in the asset management system of an operator in a province, the proportion of records with a cell azimuth angle deviation of more than 20 degrees accounted for 48.1%, which seriously affected the network. Precise operation and maintenance

After analysis, at present, operators’ community azimuth data is mainly collected by operation and maintenance personnel with hand-held compass or antenna attitude measuring instrument, and then manually entered into the operator’s asset management system. Due to the inaccuracy of the compass or antenna attitude measuring instrument and the electromagnetic waves emitted by the antenna interference, incorrect methods used, recording errors, input errors, and artificial falsification, etc., which lead to inaccurate cell azimuth data recorded in the asset management system, and each survey and verification requires driving to the designated location, which increases manpower costly and complex

Images