Summary
International Conference on Emerging Technologies for Communications
2020
Session Number:E3
Session:
Number:E3-2
Accurate Mobile Traffic Generation Scheme without Coarse-grained Data Using Conditional SR-GAN
Tomoki Tokunaga, Kimihiro Mizutani,
pp.-
Publication Date:2020/12/2
Online ISSN:2188-5079
PDF download
Summary:
Large-scale mobile traffic data analysis is an important analysis for efficiently planning mobile base station deployment plans and public transportation plans. However, since the amount of mobile traffic data increases enormously with the size and population density of the target area, its storage cost becomes much higher. To solve this problem, a scheme for generating a large amount of mobile traffic data from certain representation data has been proposed. The state-of-the-art of the scheme transforms a large amount of traffic data into a coarse-grained data and generate the original traffic data from the coarse-grained data by using Generative Adversarial Networks (GANs). However, in order to generate the original traffic data, the coarse-grained data must be preserved in storage. Consequently, the storage cost for the coarse-grained data increases as the amount of original traffic data increases. In this paper, we propose a scheme for generating a large amount of traffic data without a coarse-grained data. In concrete terms, our proposed scheme generates the original traffic data from random numbers and the target time data (label). Therefore, our scheme need not preserve a coarse-grained data. In evaluation, in order to verify the comprehensiveness of our proposed scheme, several experiments were onducted by changing the expression of random numbers and labels. In addition, we implemented the traditional scheme in the same data set with our proposed scheme. As a results, we concluded that our proposed scheme not only does reduce the storage cost for mobile traffic data by more than 20% compared to the traditional scheme, but also can generate the original mobile traffic data with a 98% accuracy.