Limitations of artificial neural networks for traffic prediction in broadband networks

Hall, James A.; Mars, P.

doi:10.1049/ip-com:20000146

Cited by 33 publications

(26 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These models can be applied to predict different types of traffic data in Ethernet, Internet, etc. ; (Shu et al, 1999); (Hall & Mars, 2000). In the following subsections, some prediction techniques are introduced including ARIMA based traffic forecasting, application of neural network in traffic forecasting, least mean square based traffic forecasting, etc.…”

Section: Traffic Prediction Techniquesmentioning

confidence: 99%

“…Neural network based traffic prediction approach is complicated to implement. The accuracy and applicability of the neural network approach in traffic prediction is limited (Hall & Mars, 2000). Artificial Neural Network (ANN) can capture the non-linear nature of network traffic and the relationship between the output and input theoretically (Hansegawa et al, 2001); (Khotanzad & Sadek, 2003); (Lobejko, 1996), however, it might suffer from over-fitting (Doulamis et al, 2003).…”

Section: Application Of Neural Network In Traffic Forecastingmentioning

confidence: 99%

See 1 more Smart Citation

Traffic Pattern Prediction Based Spectrum Sharing for Cognitive Radios

Li¹,

Zekavat²

2009

Cognitive Radio Systems

View full text Add to dashboard Cite

Section: Traffic Prediction Techniquesmentioning

confidence: 99%

Section: Application Of Neural Network In Traffic Forecastingmentioning

confidence: 99%

Traffic Pattern Prediction Based Spectrum Sharing for Cognitive Radios

Li¹,

Zekavat²

2009

Cognitive Radio Systems

View full text Add to dashboard Cite

“…With update rules defined by (14) and (15), we are able to employ AP for traffic congestion detection and prediction via clustering.…”

Section: B the Update Rules Of Affinity Propagationmentioning

confidence: 99%

“…Current work on predicting roadway traffics [4]- [13] focuses on modeling the traffic flows by analyzing the time series data. The corresponding results reveil that it is very hard to find a deterministic prediction model on general traffic flows [14], as the existing models either suffer from low prediction accuracy or only work in a particular period.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Traffic clustering and online traffic prediction in vehicle networks: A social influence perspective

Zhang

Xing

Cheng

et al. 2012

2012 Proceedings IEEE INFOCOM

View full text Add to dashboard Cite

Abstract-In this paper we investigate the dynamic traffic relationship characterized by a similarity value from one road point to another in vehicle networks. Due to the regularity of human mobility, traffic exhibits strong correlations in both temporal domain and spatial domain. By exploiting the similarity values, we derive application-specific message update rules for affinity propagation, based on which we propose an instant traffic clustering algorithm to partition the road points into time variant clusters, where the traffics within the same cluster are strongly spatially correlated. Online traffic clustering is also considered by clustering combination via evidence accumulation for further influence study. We also present a neural network based traffic prediction algorithm to predict the traffic conditions cluster by cluster for a future time based on the current and historical traffic data. Simulation study on real traffic data demonstrates that our proposed algorithms are able to identify the true influences among road points and provide accurate traffic predictions.

show abstract

Towards power consumption in optical networks: a cognitive prediction‐based technique

Sarigiannidis

Papadimitriou

Nicopolitidis

et al. 2015

Int J Communication

View full text Add to dashboard Cite

Summary Modern backbone, optical networks have developed into large, massive networks, which consist of numerous intermediate and terminal nodes as well as final users. These networks serve uninterruptedly multitude of users at the expense of considerable power consumption. Many research efforts are aimed at reducing energy consumption in large‐scale optical networks; however, this objective is deemed laborious: the operation of such networks has to continuously remain in good levels without disruptions. One of the most compelling techniques to remedy this situation is to switch off redundant links and devices at specific (short) periods. These links and devices remain idle as long as the network can cope with the underlying traffic demands. Hence, a power mechanism is required to manage how and when underutilized network elements may be silent during network operation. Nevertheless, this management entails fast processing and efficient decision making. While many research efforts neglect this serious factor, the problem of reducing the power consumption still threats the development of today's backbone network. In this work, an effective, cognitive power management technique is proposed by enhancing the decision making with traffic prediction. Traffic capacity is estimated in each link within the network supporting, thus, more efficient decisions on switching off underutilized or even idle network elements a priori. The technique introduced succeeds high accuracy levels, while it offers energy savings up to 30% lower than other energy‐aware schemes. Copyright © 2015 John Wiley & Sons, Ltd.

show abstract

Limitations of artificial neural networks for traffic prediction in broadband networks

Cited by 33 publications

References 7 publications

Traffic Pattern Prediction Based Spectrum Sharing for Cognitive Radios

Traffic Pattern Prediction Based Spectrum Sharing for Cognitive Radios

Traffic clustering and online traffic prediction in vehicle networks: A social influence perspective

Towards power consumption in optical networks: a cognitive prediction‐based technique

Contact Info

Product

Resources

About