Realtime Mobile Bandwidth Prediction Using LSTM Neural Network

Lifan Mei; Runchen Hu; Houwei Cao; Yong Liu; Zifa Han; Feng Li; Jin Li

doi:10.1007/978-3-030-15986-3_3

Realtime Mobile Bandwidth Prediction Using LSTM Neural Network

Lifan Mei, Runchen Hu, Houwei Cao, Yong Liu, Zifa Han, Feng Li, Jin Li

Electrical and Computer Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

With the popularity of mobile access Internet and the higher bandwidth demand of mobile applications, user Quality of Experience (QoE) is particularly important. For bandwidth and delay sensitive applications, such as Video on Demand (VoD), Realtime Video Call, Games, etc., if the future bandwidth can be estimated in advance, it will greatly improve the user QoE. In this paper, we study realtime mobile bandwidth prediction in various mobile networking scenarios, such as subway and bus rides along different routes. The main method used is Long Short Term Memory (LSTM) recurrent neural network. In specific scenarios, LSTM achieves significant accuracy improvements over the state-of-the-art prediction algorithms, such as Recursive Least Squares (RLS). We further analyze the bandwidth patterns in different mobility scenarios using Multi-Scale Entropy (MSE) and discuss its connections to the achieved accuracy.

Original language	English (US)
Title of host publication	Passive and Active Measurement - 20th International Conference, PAM 2019, Proceedings
Editors	Marinho Barcellos, David Choffnes
Publisher	Springer Verlag
Pages	34-47
Number of pages	14
ISBN (Print)	9783030159856
DOIs	https://doi.org/10.1007/978-3-030-15986-3_3
State	Published - 2019
Event	20th International Conference on Passive and Active Measurement, PAM 2019 - Puerto Varas, Chile Duration: Mar 27 2019 → Mar 29 2019

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11419 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	20th International Conference on Passive and Active Measurement, PAM 2019
Country/Territory	Chile
City	Puerto Varas
Period	3/27/19 → 3/29/19

Keywords

Bandwidth measurement
Bandwidth prediction
Long Short Term Memory
Multi-Scale Entropy

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-030-15986-3_3

Cite this

Mei, L., Hu, R., Cao, H., Liu, Y., Han, Z., Li, F., & Li, J. (2019). Realtime Mobile Bandwidth Prediction Using LSTM Neural Network. In M. Barcellos, & D. Choffnes (Eds.), Passive and Active Measurement - 20th International Conference, PAM 2019, Proceedings (pp. 34-47). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11419 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-030-15986-3_3

Realtime Mobile Bandwidth Prediction Using LSTM Neural Network. / Mei, Lifan; Hu, Runchen; Cao, Houwei et al.
Passive and Active Measurement - 20th International Conference, PAM 2019, Proceedings. ed. / Marinho Barcellos; David Choffnes. Springer Verlag, 2019. p. 34-47 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11419 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Mei, L, Hu, R, Cao, H, Liu, Y, Han, Z, Li, F & Li, J 2019, Realtime Mobile Bandwidth Prediction Using LSTM Neural Network. in M Barcellos & D Choffnes (eds), Passive and Active Measurement - 20th International Conference, PAM 2019, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11419 LNCS, Springer Verlag, pp. 34-47, 20th International Conference on Passive and Active Measurement, PAM 2019, Puerto Varas, Chile, 3/27/19. https://doi.org/10.1007/978-3-030-15986-3_3

Mei L, Hu R, Cao H, Liu Y, Han Z, Li F et al. Realtime Mobile Bandwidth Prediction Using LSTM Neural Network. In Barcellos M, Choffnes D, editors, Passive and Active Measurement - 20th International Conference, PAM 2019, Proceedings. Springer Verlag. 2019. p. 34-47. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-15986-3_3

Mei, Lifan ; Hu, Runchen ; Cao, Houwei et al. / Realtime Mobile Bandwidth Prediction Using LSTM Neural Network. Passive and Active Measurement - 20th International Conference, PAM 2019, Proceedings. editor / Marinho Barcellos ; David Choffnes. Springer Verlag, 2019. pp. 34-47 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{0553f9894ec34116ad80297e92b7a593,

title = "Realtime Mobile Bandwidth Prediction Using LSTM Neural Network",

abstract = "With the popularity of mobile access Internet and the higher bandwidth demand of mobile applications, user Quality of Experience (QoE) is particularly important. For bandwidth and delay sensitive applications, such as Video on Demand (VoD), Realtime Video Call, Games, etc., if the future bandwidth can be estimated in advance, it will greatly improve the user QoE. In this paper, we study realtime mobile bandwidth prediction in various mobile networking scenarios, such as subway and bus rides along different routes. The main method used is Long Short Term Memory (LSTM) recurrent neural network. In specific scenarios, LSTM achieves significant accuracy improvements over the state-of-the-art prediction algorithms, such as Recursive Least Squares (RLS). We further analyze the bandwidth patterns in different mobility scenarios using Multi-Scale Entropy (MSE) and discuss its connections to the achieved accuracy.",

keywords = "Bandwidth measurement, Bandwidth prediction, Long Short Term Memory, Multi-Scale Entropy",

author = "Lifan Mei and Runchen Hu and Houwei Cao and Yong Liu and Zifa Han and Feng Li and Jin Li",

note = "Publisher Copyright: {\textcopyright} 2019, Springer Nature Switzerland AG.; 20th International Conference on Passive and Active Measurement, PAM 2019 ; Conference date: 27-03-2019 Through 29-03-2019",

year = "2019",

doi = "10.1007/978-3-030-15986-3_3",

language = "English (US)",

isbn = "9783030159856",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "34--47",

editor = "Marinho Barcellos and David Choffnes",

booktitle = "Passive and Active Measurement - 20th International Conference, PAM 2019, Proceedings",

}

TY - GEN

T1 - Realtime Mobile Bandwidth Prediction Using LSTM Neural Network

AU - Mei, Lifan

AU - Hu, Runchen

AU - Cao, Houwei

AU - Liu, Yong

AU - Han, Zifa

AU - Li, Feng

AU - Li, Jin

PY - 2019

Y1 - 2019

N2 - With the popularity of mobile access Internet and the higher bandwidth demand of mobile applications, user Quality of Experience (QoE) is particularly important. For bandwidth and delay sensitive applications, such as Video on Demand (VoD), Realtime Video Call, Games, etc., if the future bandwidth can be estimated in advance, it will greatly improve the user QoE. In this paper, we study realtime mobile bandwidth prediction in various mobile networking scenarios, such as subway and bus rides along different routes. The main method used is Long Short Term Memory (LSTM) recurrent neural network. In specific scenarios, LSTM achieves significant accuracy improvements over the state-of-the-art prediction algorithms, such as Recursive Least Squares (RLS). We further analyze the bandwidth patterns in different mobility scenarios using Multi-Scale Entropy (MSE) and discuss its connections to the achieved accuracy.

AB - With the popularity of mobile access Internet and the higher bandwidth demand of mobile applications, user Quality of Experience (QoE) is particularly important. For bandwidth and delay sensitive applications, such as Video on Demand (VoD), Realtime Video Call, Games, etc., if the future bandwidth can be estimated in advance, it will greatly improve the user QoE. In this paper, we study realtime mobile bandwidth prediction in various mobile networking scenarios, such as subway and bus rides along different routes. The main method used is Long Short Term Memory (LSTM) recurrent neural network. In specific scenarios, LSTM achieves significant accuracy improvements over the state-of-the-art prediction algorithms, such as Recursive Least Squares (RLS). We further analyze the bandwidth patterns in different mobility scenarios using Multi-Scale Entropy (MSE) and discuss its connections to the achieved accuracy.

KW - Bandwidth measurement

KW - Bandwidth prediction

KW - Long Short Term Memory

KW - Multi-Scale Entropy

UR - http://www.scopus.com/inward/record.url?scp=85064054102&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85064054102&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-15986-3_3

DO - 10.1007/978-3-030-15986-3_3

M3 - Conference contribution

AN - SCOPUS:85064054102

SN - 9783030159856

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 34

EP - 47

BT - Passive and Active Measurement - 20th International Conference, PAM 2019, Proceedings

A2 - Barcellos, Marinho

A2 - Choffnes, David

PB - Springer Verlag

T2 - 20th International Conference on Passive and Active Measurement, PAM 2019

Y2 - 27 March 2019 through 29 March 2019

ER -

Realtime Mobile Bandwidth Prediction Using LSTM Neural Network

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this