Citywide Cellular Traffic Prediction Based on a Hybrid Spatiotemporal Network
With the arrival of 5G networks, cellular networks are moving in the direction of diversified, broadband, integrated, and intelligent networks. At the same time, the popularity of various smart terminals has led to an explosive growth in cellular traffic. Accurate network traffic prediction has beco...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-01-01
|
| Series: | Algorithms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4893/13/1/20 |
| _version_ | 1818179530823565312 |
|---|---|
| author | Dehai Zhang Linan Liu Cheng Xie Bing Yang Qing Liu |
| author_facet | Dehai Zhang Linan Liu Cheng Xie Bing Yang Qing Liu |
| author_sort | Dehai Zhang |
| collection | DOAJ |
| description | With the arrival of 5G networks, cellular networks are moving in the direction of diversified, broadband, integrated, and intelligent networks. At the same time, the popularity of various smart terminals has led to an explosive growth in cellular traffic. Accurate network traffic prediction has become an important part of cellular network intelligence. In this context, this paper proposes a deep learning method for space-time modeling and prediction of cellular network communication traffic. First, we analyze the temporal and spatial characteristics of cellular network traffic from Telecom Italia. On this basis, we propose a hybrid spatiotemporal network (HSTNet), which is a deep learning method that uses convolutional neural networks to capture the spatiotemporal characteristics of communication traffic. This work adds deformable convolution to the convolution model to improve predictive performance. The time attribute is introduced as auxiliary information. An attention mechanism based on historical data for weight adjustment is proposed to improve the robustness of the module. We use the dataset of Telecom Italia to evaluate the performance of the proposed model. Experimental results show that compared with the existing statistics methods and machine learning algorithms, HSTNet significantly improved the prediction accuracy based on MAE and RMSE. |
| first_indexed | 2024-12-11T21:05:21Z |
| format | Article |
| id | doaj.art-650cf9bfe7bd4a6f89a80a0108c6adf2 |
| institution | Directory Open Access Journal |
| issn | 1999-4893 |
| language | English |
| last_indexed | 2024-12-11T21:05:21Z |
| publishDate | 2020-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Algorithms |
| spelling | doaj.art-650cf9bfe7bd4a6f89a80a0108c6adf22022-12-22T00:50:52ZengMDPI AGAlgorithms1999-48932020-01-011312010.3390/a13010020a13010020Citywide Cellular Traffic Prediction Based on a Hybrid Spatiotemporal NetworkDehai Zhang0Linan Liu1Cheng Xie2Bing Yang3Qing Liu4School of Software, Yunnan University, Kunming 650504, ChinaSchool of Software, Yunnan University, Kunming 650504, ChinaSchool of Software, Yunnan University, Kunming 650504, ChinaSchool of Software, Yunnan University, Kunming 650504, ChinaSchool of Software, Yunnan University, Kunming 650504, ChinaWith the arrival of 5G networks, cellular networks are moving in the direction of diversified, broadband, integrated, and intelligent networks. At the same time, the popularity of various smart terminals has led to an explosive growth in cellular traffic. Accurate network traffic prediction has become an important part of cellular network intelligence. In this context, this paper proposes a deep learning method for space-time modeling and prediction of cellular network communication traffic. First, we analyze the temporal and spatial characteristics of cellular network traffic from Telecom Italia. On this basis, we propose a hybrid spatiotemporal network (HSTNet), which is a deep learning method that uses convolutional neural networks to capture the spatiotemporal characteristics of communication traffic. This work adds deformable convolution to the convolution model to improve predictive performance. The time attribute is introduced as auxiliary information. An attention mechanism based on historical data for weight adjustment is proposed to improve the robustness of the module. We use the dataset of Telecom Italia to evaluate the performance of the proposed model. Experimental results show that compared with the existing statistics methods and machine learning algorithms, HSTNet significantly improved the prediction accuracy based on MAE and RMSE.https://www.mdpi.com/1999-4893/13/1/20communication traffic predictionintelligent traffic managementdeformable convolutionattention mechanism |
| spellingShingle | Dehai Zhang Linan Liu Cheng Xie Bing Yang Qing Liu Citywide Cellular Traffic Prediction Based on a Hybrid Spatiotemporal Network Algorithms communication traffic prediction intelligent traffic management deformable convolution attention mechanism |
| title | Citywide Cellular Traffic Prediction Based on a Hybrid Spatiotemporal Network |
| title_full | Citywide Cellular Traffic Prediction Based on a Hybrid Spatiotemporal Network |
| title_fullStr | Citywide Cellular Traffic Prediction Based on a Hybrid Spatiotemporal Network |
| title_full_unstemmed | Citywide Cellular Traffic Prediction Based on a Hybrid Spatiotemporal Network |
| title_short | Citywide Cellular Traffic Prediction Based on a Hybrid Spatiotemporal Network |
| title_sort | citywide cellular traffic prediction based on a hybrid spatiotemporal network |
| topic | communication traffic prediction intelligent traffic management deformable convolution attention mechanism |
| url | https://www.mdpi.com/1999-4893/13/1/20 |
| work_keys_str_mv | AT dehaizhang citywidecellulartrafficpredictionbasedonahybridspatiotemporalnetwork AT linanliu citywidecellulartrafficpredictionbasedonahybridspatiotemporalnetwork AT chengxie citywidecellulartrafficpredictionbasedonahybridspatiotemporalnetwork AT bingyang citywidecellulartrafficpredictionbasedonahybridspatiotemporalnetwork AT qingliu citywidecellulartrafficpredictionbasedonahybridspatiotemporalnetwork |