Citation: Tokunaga, T.; Mizutani, K.
A Comprehensive Evaluation of
Generating a Mobile Traffic Data
Scheme without a Coarse-Grained
Process Using CSR-GAN. Sensors
2022, 22, 1930. https://doi.org/
10.3390/s22051930
Academic Editor: Shih-Chia Huang
Received: 22 December 2021
Accepted: 28 February 2022
Published: 1 March 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
A Comprehensive Evaluation of Generating a Mobile
Traffic Data Scheme without a Coarse-Grained Process
Using CSR-GAN
†
Tomoki Tokunaga
1,‡
and Kimihiro Mizutani
1,2,
*
,‡
1
Graduate School of Science and Engineering, Kindai University, 3-4-1 Kowakae,
Higashiosaka 577-0818, Osaka, Japan; 2033340424m@kindai.ac.jp
2
Cyber Informatics Research Institute, Kindai University, 3-4-1 Kowakae,
Higashiosaka 577-0818, Osaka, Japan
* Correspondence: mizutani@info.kindai.ac.jp
† This paper is an extended version of our paper published in Tokunaga, T.; Mizutani, K. A scheme of
estimating mobile traffic data without coarse-grained process using conditional SR-GAN. IEICE Commun.
Express 2021, 10, 441–446. https://doi.org/10.1587/comex.2021ETL0017.
‡ These authors contributed equally to this work.
Abstract:
Large-scale mobile traffic data analysis is important for efficiently planning mobile base
station deployment plans and public transportation plans. However, the storage costs of preserving
mobile traffic data are becoming much higher as traffic increases enormously population density of
target areas. To solve this problem, schemes to generate a large amount of mobile traffic data have
been proposed. In the state-of-the-art of the schemes, generative adversarial networks (GANs) are
used to transform a large amount of traffic data into a coarse-grained representation and generate the
original traffic data from the coarse-grained data. However, the scheme still involves a storage cost,
since the coarse-grained data must be preserved in order to generate the original traffic data. In this
paper, we propose a scheme to generate the mobile traffic data by using conditional-super-resolution
GAN (CSR-GAN) without requiring a coarse-grained process. Through experiments using two real
traffic data, we assessed the accuracy and the amount of storage data needed. The results show
that the proposed scheme, CSR-GAN, can reduce the storage cost by up to 45% compared to the
traditional scheme, and can generate the original mobile traffic data with 94% accuracy. We also
conducted experiments by changing the architecture of CSR-GAN, and the results show an optimal
relationship between the amount of traffic data and the model size.
Keywords: conditional GAN; SR-GAN; traffic data management
1. Introduction
Mobile traffic data have increased rapidly in recent years due to the dramatic spread
of mobile devices and online services. According to Ericsson research, the amount of
monthly mobile traffic data is expected to reach 49 EB per month by the end of 2020 and
237 EB by 2026 [
1
]. To handle a large amount of future mobile traffic data, it is important
to analyze the trends of mobile traffic patterns in order to deploy mobile devices and
online services on a large scale in urban areas [
2
–
5
]. However, the computational power to
achieve this is high [
6
], since specialized equipment (e.g., measurement probes) is required.
Furthermore, the storage cost of mobile traffic data is becoming much higher, since it
increases enormously with the size and population density of a target area. In order to
solve these problems, schemes to estimate a large amount of traffic data have been proposed.
In particular, with the recent advances in neural network technology, many schemes using
deep neural networks have been widely proposed.
For example, recurrent neural networks (RNNs) used in LSTM architecture [
7
] are
superior to traditional machine learning approaches in terms of accurate traffic generation.
Sensors 2022, 22, 1930. https://doi.org/10.3390/s22051930 https://www.mdpi.com/journal/sensors
