Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 4 de 4
Filter
1.
Computer Networks ; : 109518, 2022.
Article in English | ScienceDirect | ID: covidwho-2149583

ABSTRACT

The Covid-19 pandemic has forced the workforce to switch to working from home, which has put significant burdens on the management of broadband networks and called for intelligent service-by-service resource optimization at the network edge. In this context, network traffic prediction is crucial for operators to provide reliable connectivity across large geographic regions. Although recent advances in neural network design have demonstrated potential to effectively tackle forecasting, in this work we reveal based on real-world measurements that network traffic across different regions differs widely. As a result, models trained on historical traffic data observed in one region can hardly serve in making accurate predictions in other areas. Training bespoke models for different regions is tempting, but that approach bears significant measurement overhead, is computationally expensive, and does not scale. Therefore, in this paper we propose TransMUSE (Transferable Traffic Prediction in MUlti-Service Edge Networks), a novel deep learning framework that clusters similar services, groups edge-nodes into cohorts by traffic feature similarity, and employs a Transformer-based Multi-service Traffic Prediction Network (TMTPN), which can be directly transferred within a cohort without any customization. We demonstrate that TransMUSE exhibits imperceptible performance degradation in terms of mean absolute error (MAE) when forecasting traffic, compared with settings where a model is trained for each individual edge node. Moreover, our proposed TMTPN architecture outperforms the state-of-the-art, achieving up to 43.21% lower MAE in the multi-service traffic prediction task. To the best of our knowledge, this is the first work that jointly employs model transfer and multi-service traffic prediction to reduce measurement overhead, while providing fine-grained accurate demand forecasts for edge services provisioning.

2.
Journal of Web Engineering ; 21(5):1419-1433, 2022.
Article in English | Web of Science | ID: covidwho-1998051

ABSTRACT

To fix network congestion resulting from the increase in high volume traffic in data-intensive science and the increase in internet traffic due to COVID19, there has been a necessity of traffic engineering through traffic prediction. For this, there have been various attempts from a statistical method such as ARIMA to machine learning including LSTM and GRU. This study aimed to collect and learn KREOENT backbone and subscribers' traffic volume through diverse machine learning techniques (e.g., SVR, LSTM, GRU, etc.) and predict maximum traffic on the following day.

3.
14th International Conference on COMmunication Systems and NETworkS, COMSNETS 2022 ; : 812-817, 2022.
Article in English | Scopus | ID: covidwho-1722903

ABSTRACT

Traffic prediction and analysis is an essential task towards intelligent mobility, particularly for path planning and navigation. When the traffic flow starts after the COVID-19 pandemic is subsided, the mobility patterns changes and may become unpredictable or challenging. This problem may be crucial, particularly if many people hurry to single occupancy transport mode. Notably, the rapid development in machine learning with new methods and the emergence of new data sources make it possible to evaluate and predict traffic conditions in smart cities more quickly and precisely. The proposed work is modeled in two-fold manner to investigate the impact of COVID shift in regular urban traffic movements given the particular period of the pre, during, and post lockdown phases. Firstly, the investigation is carried out for time series analysis considering the three phases of lockdown. Secondly, the real-time spatial information is analyzed for different time zones in a day. Notably, this requires a detailed analysis of the heterogeneous and complex input traffic data. Machine learning and advanced deep learning methodologies such as regression models, RNN, variants of LSTM, and GRU is used for analysis in this proposed traffic modeling. Significantly, the least error scores with Root Mean Square Error (RMSE) loss of 1.82 is observed for the RNN and GRU models, and 0.058 with the Gradient Boosting regression analysis, respectively. © 2022 IEEE.

4.
EAI/Springer Innovations in Communication and Computing ; : 163-178, 2022.
Article in English | Scopus | ID: covidwho-1627268

ABSTRACT

Writing software (programs) is an obstruction, we do not have so many good developers who can develop much more enhanced models and so, for this purpose today many use the data instead of people to perform the same task. According to the generations’ needs, the programmers developed the machine learning approach to make the programming much more scalable and expandable in this domain. Before, machine learning traditional programming is a much more famous approach where programmers used to code each and every single line with their own and its main drawback is that it is not so much scalable. Here, in this chapter we are going to discuss various applications of machine learning and the algorithms they are using along with their advantage, disadvantage, and its working model of how much the particular application is scalable. Like we are going to discuss virtual personal assistants, email spam, online fraud detection, traffic predictions, social media personalization, and many more. In coming to algorithms, we will get to know about Naive Bayes algorithms, neural networks, KNN algorithms, linear regression model, logistic regression model, etc. On coming to today’s need we are also going to discuss its applications in detection of COVID-19 defaulters by the use of semantic segmentation algorithms. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

SELECTION OF CITATIONS
SEARCH DETAIL