Wireless cell information detection method based on semi-supervised clustering

Abstract

The present invention is a wireless cell information detection method based on semi-supervised clustering, which aims to solving the problem of how to acquire the key parameters of the base stations and cells of a mobile network timely and accurately as well as the operating status of the base stations so as to construct and update the base station information table and to make the operators better grasp the deployment and operation of base stations in the network for effective management. The method comprises: inputting the on-net information sampling data acquired by the users' mobile terminals; Step S1: terminal data cleaning; Step S2: station type determining; Step S3: initial positioning of the cell sites using the vector clustering algorithm; Step S4: cell site address aggregating in the same base station; Step S5: cell angles and direction angles determining using a soft-interval SVM support vector machine; and Step S6: adding fields such as cell states and change dates when updating the base station information table used to indicate which cells have been changed at what time (newly built, removed, relocated, parameter updated, etc.).

Description

Background technique [0001] For a mobile network, a cell (also referred to as a sector, cell) is a basic unit for dividing a network coverage area. The base station information table is the core basic data required by telecom operators for network operations. It describes the basic parameters of all cells under a network, generally including station name, cell name, site longitude and latitude, cell identity, station type, direction angle, Pitch angle, station height, coverage scene, etc. The cell identification uses different identification methods for different network standards. For example, a cell in a GSM / WCDMA / TDSCDMA network is composed of two parameters, LAC and CI, to determine a unique cell in the network. For an LTE network, then A cell is uniquely determined by a combination of three parameters: TAC, eNodeBID, and CellID. The station type means that the base station to which the cell belongs is an outdoor omnidirectional station (only one cell), an outdoor direct...

AI Technical Summary

Problems solved by technology

[0007] (1) A large number of manual participation may lead to possible data errors

[0008] (2) There is a large time delay in reporting level by level, and the timeliness of information in the base station information table cannot be guarant

Images