Network availability prediction with hidden context

Rathnayake, Upendra; Ott, Maximilian; Seneviratne, Aruna

doi:10.1016/j.peva.2011.03.003

Cited by 5 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then we have shown that Dynamic Baysean Networks based Prediction mechanism can be combined with a report based information exchange scheme to develop such a network predication system. We also provided a summary of our own Dynamic Baysean Networks based Prediction mechanism [23] and report exchange system [16] which on their own provide good predication. Therefore we are confident that a hybrid scheme would provide even better and more importantly robust network availability predication system.…”

Section: Discussionmentioning

confidence: 99%

“…We have shown in our earlier work how to find the optimal thresholds for variables such as the Bluetooth devices around using a conditional independence check [23]. In addition, it was found that the variable, whether the LAN cable is plugged in, did not contribute to the prediction accuracy in general.…”

Section: A Dbn Based Predictionmentioning

confidence: 98%

“…With this scheme, uses the reports described in [16] to augment the local measurements for the determination of context variables described. Then use a Dynamic Baysean Network with hidden variables as described in [23] to predict the performance.…”

Section: Hybrid Systemmentioning

confidence: 99%

“…In the following sections we summarise some of our previous work [23], [16] which show the viability of the proposed scheme. …”

Section: Fig 1: Proposed Hybrid Systemmentioning

confidence: 99%

“…This enabled the use of the time of the day variable implicitly, thus reducing the reliance to other four observable variables. The prediction accuracy using this information and DBN as described in [23], as shown in Figure 3. As can be seen, it is possible to obtain a prediction accuacry of approximately 70% even in the worst case.…”

Section: A Dbn Based Predictionmentioning

confidence: 99%

See 4 more Smart Citations

Network availability prediction: Can it be done?

Seneviratne

Pedrasa

Rathnayake

2011

Global Information Infrastructure Symposium - GIIS 2011

Self Cite

View full text Add to dashboard Cite

With the availability of more powerful mobile devices and a variety of access networks, users are expecting more and more services whilst on the move. There have been many attempts develop methods of providing these types of services form a user as well as a service provider perspective. All of these methods are based on the ability to predict the future. In this paper, will first present an overview of the research on one aspect of this, namely network availability prediction. We first, summarise the work that has been done in network availability prediction and categorize them. Using the categorisation, we show that one of the existing mechanisms provide the necessary accuracy and robustness. Then we present a hybrid design which overcome the limitations of the current systems. We show the viability of the the proposed hybrid system by summarising a dynamic baysean network and report exchange based predication mechanisms. We conclude the paper with a brief discussion on the open issues of developing such a hybrid scheme.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: A Dbn Based Predictionmentioning

confidence: 98%

Section: Hybrid Systemmentioning

confidence: 99%

“…In the following sections we summarise some of our previous work [23], [16] which show the viability of the proposed scheme. …”

Section: Fig 1: Proposed Hybrid Systemmentioning

confidence: 99%

Section: A Dbn Based Predictionmentioning

confidence: 99%

See 3 more Smart Citations

Network availability prediction: Can it be done?

Seneviratne

Pedrasa

Rathnayake

2011

Global Information Infrastructure Symposium - GIIS 2011

Self Cite

View full text Add to dashboard Cite

show abstract

Vertical Handover Decision Algorithm for Multimedia Streaming in VANET

Hassoune¹,

Maaza²,

Senouci

2017

Wireless Pers Commun

View full text Add to dashboard Cite

International audienceVehicular ad hoc networks represent one of the most important applications of wireless ad hoc networks. In highways, where different networks exist, frequent and sometimes unnecessary switches from one network to another may occur, which degrade network performance and affect applications, especially those that require high quality networks such as multimedia applications (Video conferencing, Video On-Demand, etc.). This work presents an analytical model and a Vertical Handover decision method for Highways called VHH. It is based on position, velocity, jitter, and density as mandatory inputs which aim to both minimize Vertical handover frequency and avoid unnecessary handoffs and ping pong effect between different networks, in the goal of enhancing multimedia streaming services in highways. A simulation is provided to prove the performance of the algorithm

show abstract