Year-end Sale % OFF 
Abstract
For the network service construction and optimization of wireless cell, the effective scene division is an important basis for formulating more accurate network construction schemes and optimization strategies. The traditional cell scene division method is manually divided according to the single-dimensional business indicators, but there are some problems such as the inaccuracy of division and the inability to visualize. In this paper, we propose a cell scene division and visualization method based on autoencoder and K-means algorithm. We train an autoencoder network to conduct the dimension reduction of the wireless perception key quality indicator (KQI) data of cells, and then use elbow method and K-means algorithm to cluster the dimension-reduced data precisely. Through statistical analysis and comparison of indicators of cells in different classes obtained by clustering, we finally achieve accurate cell scene division and visualization.
Highlights
With the rapid development of mobile communication network [1]–[8], the network service quality of wireless cell has gradually become a key factor in the core competitiveness of communication operators
As the number of mobile communication users and the amount of business increase, different business in the network is distributed in different wireless cells, which leads to different resource allocation requirements of wireless cells
We propose a method based on the autoencoder and K-means algorithm to achieve the cell scene division
Summary
With the rapid development of mobile communication network [1]–[8], the network service quality of wireless cell has gradually become a key factor in the core competitiveness of communication operators. The traditional cell scene division is mainly based on the single-dimension business indicator and manually divided according to the experience. This kind of method is a coarse-grained qualitative division, which cannot consider the occupancy of resources by cell business in an all-round way, so it cannot be used as an accurate basis for adjustment and optimization, nor can it guide network construction and capacity expansion adjustment. We propose a method based on the autoencoder and K-means algorithm to achieve the cell scene division. We firstly use the autoencoder network to conduct the dimension reduction of the wireless perception KQI data of cells, and determine the real clustering number of the dimensionreduced data by the elbow method.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
