2016 IEEE 10th International Conference on Application of Information and Communication Technologies (AICT) 2016
DOI: 10.1109/icaict.2016.7991765
|View full text |Cite
|
Sign up to set email alerts
|

Monitoring of qualitative changes of network traffic states based on the heteroscedasticity effect

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

1
1
0

Year Published

2018
2018
2021
2021

Publication Types

Select...
1
1

Relationship

1
1

Authors

Journals

citations
Cited by 2 publications
(2 citation statements)
references
References 6 publications
1
1
0
Order By: Relevance
“…The obtained results allow to state that the established gamma distribution parameters allow to ensure the adequacy of the simulation model. The simulation results for the gamma distribution confirm the stable detection regions of the G-effect obtained earlier [6] for the normal law of the distribution of requests' intensities. Moreover, in the model, based on the gamma distribution of the hetero regions, they have a bigger time length than the model using the normal distribution law.…”
Section: Experimental Researches Of Qualitative Change In Technologicsupporting
confidence: 84%
“…The obtained results allow to state that the established gamma distribution parameters allow to ensure the adequacy of the simulation model. The simulation results for the gamma distribution confirm the stable detection regions of the G-effect obtained earlier [6] for the normal law of the distribution of requests' intensities. Moreover, in the model, based on the gamma distribution of the hetero regions, they have a bigger time length than the model using the normal distribution law.…”
Section: Experimental Researches Of Qualitative Change In Technologicsupporting
confidence: 84%
“…At this stage it is assumed to use the principal component and multidimensional scaling methods. At the stage of "data analysis", hidden insights are supposed to be identified by the mutual correlation (quantitative and qualitative) of the user's characteristics of the media environment [13][14]. To do this, we plan to use the canonical correlation method.…”
Section: Figure 1: Classification Of Influence Modelsmentioning
confidence: 99%