A Tracking Area Planning Method Based on Spectral Clustering

Abstract

The present invention proposes a tracking area planning method based on spectrum clustering, which is mainly aimed at the user tracking area planning method in the small cell network environment of the hotspot area. The overall idea of ​​the method proposed by the present invention is, first, a system model is constructed based on the small cell deployment of the Poisson point process, and a cellular network diagram showing user mobility and paging characteristics is generated; then, the TA planning problem is modeled as a graph correlation coefficient The segmentation problem; and then apply the spectral clustering algorithm based on graph theory to carry out TA planning on the generated network graph.

Description

technical field [0001] The present invention relates to the technical field of user location management in the communication field, and mainly relates to a planning method for a tracking area (Tracking Area, TA). Background technique [0002] In recent years, the rapid increase in the number of mobile users has led to a rapid increase in mobile communication network traffic, which greatly increases the capacity requirements of the network. Large hot spots, such as shopping malls, have a large number of connected devices. The dense deployment of small cell base stations is imminent, and ultra-dense networking has emerged as the times require. The future small cell network will deploy dense self-organizing, low-cost, low-power small cells in hot spots within the range of macro cells. Although the small cell network has many advantages, due to the characteristics of dynamic random deployment, ultra-dense, strong self-organization and liberalization capabilities of small cells...

AI Technical Summary

Problems solved by technology

In 2013, Toril et al. calculated the mobile characteristics of mobile users in the TA planning area, modeled the TA planning problem as a graph segmentation problem, and used the evolutionary search algorithm to give a TA planning solution. This algorithm belongs to the latest heuristic TA planning algorithm, but , the algorithm steps are complex, and the speed of searching for the global optimal solution of the TA planning problem is slow. In the face of the ultra-dense network of small cell deployment environment, it is necessary to study a more efficient TA planning algorithm

In 2017, Ning et al. proposed a TA planning algorithm based on community detection, which modeled the TA planning problem as a community detection problem in a complex network, and applied a cooperative game-based community detection algorithm to give a TA planning solution. When there are many small cells, it will lead to an unreasonable phenomenon that more and more small cells are divided into one TA

In addition, most of the current research on dynamic location management methods is based on regular cellular topology models with the same size, shape, and distribution of cells, which cannot cope with the ultra-dense networking of cellular deployment environments in hotspot areas.

In addition, with changes in user mobility trends and cellular base station deployments, the initial TA planning method cannot better optimize the signaling cost overhead of the network system

Images