SDRANK: ADAPTABLE SERVICE SELECTION FOR IOT BASED ON RANKING

• Main Author: CHRISTOPER KAKUNSI (NIM: 23516047), DEDDY
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/26444
• Institution: Institut Teknologi Bandung

<p align="justify"> Internet of Things (IoT) is a computing paradigm that integrate the real-world object to the existing virtual world and enables remote interaction through the intenet using standard mechanism. The development of networking technology and increasing in device’s computing capacity drives the number of connected objects. This number increase rapidly and requires efficient service management. IoT system could be developed using service-oriented and event-driven paradigm. One of the challenges in developing service-oriented IoT system is selecting the most relevant service and adapting dynamicity of IoT devices. <br /> <br /> This research proposes an IoT service model and IOT-WSDL, and its implementation in SDRank system, an adaptable service selection method using ranking. IoT service model developed based on existing researches on IoT context, while IOT-WSDL extended from WSDL 2.0. IoT services could be described dan sorted based on the models to generate ranking. The implementation of ranking-based service selection involves two method, which is Service Rating, based on utility function and Analytical Hierarchy Process (AHP). The generated ranking could be used selecting services and adapting changes. <br /> <br /> The result shows that ranking approach can be used in IoT service selection with concern of the dynamicity of IoT devices. The evaluation of both methods shows that ranking generated by AHP is more stable, thus it is more suitable for selection with ranking stability as the primary requirement. While, rating perform better than AHP, thus it will be the first choice when response time is crucial. The response time of both method is different but not significant. <p align="justify">