Abstract
AbstractIn pervasive computing systems, service selection and fault tolerance are two primary factors, which determines reliability and efficient of service selection system. In this paper, according on ANN-based (Artificial Neural Network) service selection model, we extend the definition and formal description of service, and propose the architecture of service migration system, evaluation method, judgment theorems of triggering and a serviceoriented seamless migration (SOSM) algorithm in pervasive computing systems. We also propose the flexible communication protocol for service seamless migration activity based on mobile agent computing network. We have implemented the SOSM mechanism and flexible communication protocol in an intelligent service selection prototype (ISSPS) system, which provide the pervasive web services for users in pervasive computing environment. The results show that the proposed service migration mechanism is not only reliable but also efficient.
Highlights
The dawn of the 21st century has seen explosive interest in Pervasive/Ubiquitous Computing
We want to implement an example scenario: a traveller arriving at one of webservice-bars finds a barcode service token, and her mobile device reads the token through its inbuilt camera and decodes the service identifier, and he can select all kinds of pervasive web services that servers can provide, and when the selected service provider can not provide the service of satisfying QoS requests, the service program and data are migrated to a new service provider with the same or similar function and the service migration process is transparent to user, namely users do not have to handle details of service migration activity and only concern with the results of service and QoS
If the percent of successful service activity is below 80%, it is considered that the performance of service migration system is not eligible, i.e. the threshold of jitter error is set to no more than 0.2
Summary
The dawn of the 21st century has seen explosive interest in Pervasive/Ubiquitous Computing. It represents seamless migration at the granularity of individual processes, and has been a research focus on many experimental operating systems. In a dynamic and heterogeneous environment (e.g. due to disconnections or mobility etc.), the network service provided to users may change and may no longer meet the application requirements. This needs to research the problem of service seamless migration by adapting the networking environment, or the application, or its communication streams, or using a combination of these different types of adaptations. We put forward the architecture of service migration system and a service-oriented intelligent seamless migration algorithm
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
