Clustering Method for Mobile Traffic Prediction 


Vol. 47,  No. 2, pp. 398-407, Feb.  2022
10.7840/kics.2022.47.2.398


PDF
  Abstract

In order to accommodate mobile traffic and maintain network performance, it is important to predict future mobile traffic generation. In this paper, we predict the amount of future traffic data using convolutional LSTM (convLSTM) that is adequate deep learning space-time data modeling. Because traffic data have different attribute over time and space, the learning with all traffic data with different characteristics at once can degrade the performance of the model. Therefore, in this paper, we set the criteria for distinguishing traffic similarity, and group the dataset using the clustering algorithm that reflects the selected criteria. Then, the deep learning model learns traffic data on a cluster-by-cluster basis. This paper describes the similarity-based clustering method and analyzes the traffic prediction performances as the number of clusters is increased. The learning results show that as the number of clustering increases, prediction errors decrease. However, too many clustering results in increase of prediction errors.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

S. Na, Y. Kim, H. You, H. Ahn, J. Moon, E. Hong, "Clustering Method for Mobile Traffic Prediction," The Journal of Korean Institute of Communications and Information Sciences, vol. 47, no. 2, pp. 398-407, 2022. DOI: 10.7840/kics.2022.47.2.398.

[ACM Style]

Se-Hyeon Na, Young-Jun Kim, Hyeon-Min You, Hee-Jun Ahn, Jung-Mo Moon, and Een-Kee Hong. 2022. Clustering Method for Mobile Traffic Prediction. The Journal of Korean Institute of Communications and Information Sciences, 47, 2, (2022), 398-407. DOI: 10.7840/kics.2022.47.2.398.

[KICS Style]

Se-Hyeon Na, Young-Jun Kim, Hyeon-Min You, Hee-Jun Ahn, Jung-Mo Moon, Een-Kee Hong, "Clustering Method for Mobile Traffic Prediction," The Journal of Korean Institute of Communications and Information Sciences, vol. 47, no. 2, pp. 398-407, 2. 2022. (https://doi.org/10.7840/kics.2022.47.2.398)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880