Akademik Veri Yönetim Sistemi
Tezin Türü: Yüksek Lisans
Tezin Yürütüldüğü Kurum: Işık Üniversitesi, Fen Bilimleri Enstitüsü, Bilgisayar Mühendisliği, Türkiye
Tezin Onay Tarihi: 2019
Tezin Dili: İngilizce
Öğrenci: Cem Türker
Danışman: Ayşegül Tüysüz Erman
Özet:
In recent years, Software De ned Networking is started to shape telecommunication industry. Control plane and data plane separation is applied on telecommunication infrastructure and have evolved packet network. There are also many studies about the application of Software De ned Networking on Wireless Sensor Networks domain. One of them is SDN-WISE that aims to reduce the complexity of network con guration and management of Wireless Sensor Networks. In this thesis, a couple of contributions are done on SDN-WISE architecture to improve the network performance by changing the routing mechanism. In wireless sensor networks, nodes, who are close to sink, places a critical role, because those nodes are being frequently selected to reach destination path when a data is required to deliver on edge nodes in the topology. In this thesis, we try to maximize the lifetime of the network by selecting destination paths considering forecasted remaining battery level. Lifetime de ned as follows. When any node dies in a topology, it is assumed that the lifetime of the network is over. This approach is implemented on SDN-WISE architecture and to see the performance of this method a set of experiments are performed. This thesis is about maximizing the lifetime of the network on SDN-WISE by applying forecasting on node's battery level. Architectural modi cations is doneand linear regression is added to reach the objective. Our results show that the lifetime of the network is increased. However, while increasing the lifetime of the network, the packet delivery delay is increased between the nodes. We analyse this trade-off.