A Routing Protocol for Software-Defined Underwater Internet of Things Using the Minimum-Cost Flow Model

نوع: Type: Thesis

مقطع: Segment: masters

عنوان: Title: A Routing Protocol for Software-Defined Underwater Internet of Things Using the Minimum-Cost Flow Model

ارائه دهنده: Provider: Seif Al-Din Saddam Hussein Fathi

اساتید راهنما: Supervisors: Dr. Reza Mohammadi

اساتید مشاور: Advisory Professors:

اساتید ممتحن یا داور: Examining professors or referees: Dr. Muharram Masorizade - Dr. shakoor Vakilian

زمان و تاریخ ارائه: Time and date of presentation: 2026

مکان ارائه: Place of presentation: seminar

چکیده: Abstract: The Internet of Underwater Things requires routing methods tailored to underwater environments because of the characteristics of acoustic channels, including high propagation delay, limited bandwidth, packet loss, and restricted node energy. Software-defined networking separates the control plane from the data plane and provides a centralized view of the network, thereby enabling more coherent routing decisions for data transmission. This research aims to design and implement a routing protocol for the Software-Defined Internet of Underwater Things using the minimum-cost flow model. In the proposed method, the software-defined controller receives information about the network topology, link conditions, and residual energy of the nodes and models the network as a directed graph. The cost of each arc is calculated by combining distance-dependent transmission energy with a penalty for insufficient residual energy at the destination node. After removing nodes and arcs with inadequate capacity, one unit of flow is sent from each source to the logical destination, and the lowest-cost path is determined. The selected path is installed in the network as forwarding rules and is periodically recalculated according to the updated state of the nodes. The proposed method was evaluated in a discrete-event simulator using 25 scenarios obtained from combinations of five packet transmission rates and five packet sizes. In addition, 225 records associated with nine methods and 1,800 numerical values from the project outputs were audited and verified against the final results file. The findings confirmed the correct operation of the algorithm in constructing the graph, calculating arc costs, removing dead nodes, determining valid paths, and installing forwarding rules. For the minimum-cost flow method, the mean derived packet delivery indicator was 0.9585, while the mean Average Loss was 0.0415. Compared with the best corresponding baseline method, Average Loss was reduced by an average of 89.38%. However, because this metric is defined in a nonstandard manner, this finding is considered conditionally positive and should not be interpreted as a direct packet delivery ratio. In contrast, the recorded energy consumption of the proposed method was, on average, 3.58 times that of the best baseline method. The mean number of dead nodes reached 12.8, and the average delay penalty relative to the fastest method was approximately 4.10%. Therefore, the hypotheses concerning reduced energy consumption and increased network lifetime were not supported by the available data. Overall, the main contributions of this research are an implementable software-defined architecture and minimum-cost flow model, the generation of complete and auditable results, and the identification of the trade-off among delivery capability, energy consumption, and delay. Although the proposed framework successfully performs the routing procedure, achieving the intended energy and network-lifetime objectives requires revising the loss metric, integrating the energy models, normalizing the cost function, and conducting repeated experiments