A novel intelligent service selection algorithm and application for ubiquitous web services environment
Introduction
In 1991, Mark Weiser put forward ubiquitous/pervasive computing which integrates computation into the environment, rather than having computers which are distinct objects. Promoters of this idea hope that ubiquitous computing embedding the integration of computer, communication and digital media technology makes it possible to integrate the physical world we are living in and the virtual world in the information space together as the whole, which would enable people to move around and interact with computers more naturally than they currently do (Weiser, 1991, Weiser, 1993). The function of single embedded system is so perfect that almost all local tasks can be resolved in existed systems. But because of increasing of computing complexity and mobility for ubiquitous computing environment, more and more task cannot be finished by a single device. So there must be more than one device to cooperate to finish the target task. Therefore, how to transparently offer ubiquitous computing service to users by hiding heterogeneous architecture of underlying software and hardware infrastructure and how to choose a suitable service from all the useable services regardless of user’s location are most important steps in ubiquitous computing domain (Holmquist et al., 1993, Weiser, 1993).
In recent years, many studies have focused on developing feasible mechanisms and systems to select appropriate service from service resources systems in order to improve performance and efficiency. Many researchers have proposed all kinds of solutions to the problem of service selection such as the Agent scheme (Inverno et al., 1994, Lan et al., 1999, Shehory, 1999), the Jini scheme (Sun company, 2003, Wenying et al., 2003) and the UPnP scheme (Michael & Jack, 1987), etc. Other researchers improved the service selection scheme based on trust-mechanism (Marco & Alberto, 1992). However, in these traditional or improved models there are not an effective guidance provided to users and they don’t effectively consider that the service selection activity is grievously affected by implicit context information, so the service selection process is out of control. It can be the result of low system performance and bad efficiency. Ubiquitous computing services aim at exploiting the full range of sensors and network available. A core characteristic of ubiquitous computing systems is that they are context-aware, in the sense that they are able to provide services not only based on information that end users provide, but also based on implicit contextual information (Dey, 2001). Implicit information is usually derived based on a rich collection of casually accessible, often invisible sensors that are connected to a network structure. Apart from context-awareness, ubiquitous computing systems feature increased dynamism and heterogeneity, which differentiate them radically from traditional distributed system. The underlying ubiquitous computing infrastructures are more sophisticated and bring into foreground issues such as user mobility, disconnection, dynamic introduction and removal of devices, diverse network connections, as well as the need to blend the physical environment with the computing infrastructure (Murphy, Picco, & Roman, 2001).
Therefore, a novel service selection algorithm based on Artificial Neural Network (called the ANNSS algorithm) for ubiquitous computing environment is proposed after study the existing service selection algorithm deeply and discard the improved service selection algorithm based on genetic algorithm, fuzzy logic or trust-mechanism. In new method, the traditional service selection model was reduced and abstracted the most important factors to construct new model. According to the earlier information of the cooperation between the devices and the context information according to their occurrence times in ubiquitous computing system, an ANN-based evaluation standard for the service quality of service provider is given out so that user can acquire an effective guidance and choose a most appropriate service from many service providers. In order to satisfy the requirements of time issue in real-time system, we improved the traditional BP algorithm based on three-term. We have fulfilled various simulations in an actual ubiquitous web services system and the results of simulation show that the proposed control scheme is not only scalable but also efficient. The service selection control scheme with ANNSS superior to the traditional service selection method without ANNSS, and that the novel BP algorithm based on three-term has high convergence speed and good convergence stability. The novel method can exactly choose a most appropriate service from many service providers and provide the most perfect service performance to users.
The structure of the paper is as follows. Section 2 formulates the definition and formalized specification for service. Section 3 describes main results including the novel ANN-based service selection model, process of service selection algorithm and Event-condition-action Mechanism, a context model, evaluation system and improved BP algorithm based on three-term used in our solution. Section 4 describes application of ANNSS algorithm in an adopted prototype, and discusses experiment results. Section 5 summarizes the paper and outlines the main conclusions.
Section snippets
Definition of service
Within the service architecture for ubiquitous computing environment, service is the most important conception. Services can be implemented as either hardware devices, software programs, or a combination of the two, and can be found by human and computational clients. The definition of service in existed distributed system is appropriate for ubiquitous computing environment no longer because:
- (a)
The connotation of service is different from each other. In ubiquitous computing system, a service is an
ANN-based service selection model
Our model of service selection control model based on ANN is shown in Fig. 2. The heart of the service system is a quartet of protocols called discovery, join, lookup and evaluation. A pair of these protocols-discovery and join-occur when a device is plugged in. Discovery occurs when a service is looking for a lookup service and which to register. Join occurs when a service has located a lookup service and wishes to join it. Lookup occurs when a client or user needs to locate and invoke a
Experiments environment and objectives
In order to evaluate the performance of above proposed ANN-based service selection algorithm, we have implemented the ANNSS algorithm in an ubiquitous web services system shown in Fig. 5 and fulfilled all kinds of simulations. We want to implement a example scenario: a traveller arriving at one of web-service-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 ubiquitous web
Conclusion
This paper proposed a novel ANN-based service selection algorithm (i.e. ANNSS algorithm) and a improved BP algorithm based on three-term consisting of learning rate (LR), a momentum factor (MF) and a proportional factor (PF). The ANNSS algorithm is scalable because it can dynamically evaluate Qos of service provider according to the earlier information of the cooperation between the devices and the context information for ubiquitous computing environment. The improved BP algorithm can is
Acknowledgements
The authors would like to thank the support of the Ministry of Education Technology Research Key Foundation of China under Grant No. 104086.
References (25)
- et al.
On the development of a web-based system for transportation services
Information Sciences
(2006) - Chen, H., Finin, T., & Joshi, A. (2003). An ontology for context-aware pervasive computing environments. Special Issue...
- Christensen, E., Curbera, F., Meredith, G., & Weerawarana, S. (2001). Web Services Description Language(WSDL)1.1, W3C...
Approximations by superpositions of a sigmoidal function
Math Control Signal System
(1989)Understanding and using context
Personal and Ubiquitous Computing Journal
(2001)- et al.
Design of a trust valuation model in software service coordination
Journal of Software
(2003) Theory of back propagation neural network
Proceedings of IJCNN
(1989)- et al.
Combining subjective and objective QoS factors for personalized web service selection
Expert Systems with Applications
(2007) - et al.
Supporting group collaboration with inter-personal awareness devices
Personal Technologies
(1993) - et al.
Collaborative interface agents
An introduction to computing with neural net
IEEE ASSP Magazine
Cited by (29)
Flexible user-centric service selection algorithm for Internet of Things services
2014, Journal of China Universities of Posts and TelecommunicationsRepresentation and reasoning of context-dependant knowledge in distributed fuzzy ontologies
2010, Expert Systems with ApplicationsCitation Excerpt :Not surprisingly, ontologies have been proposed to be used for modeling context knowledge, as they provide some advantages over other formalisms: reusability, sharing, reasoning, standardization, supporting tools, etc. (Bobillo et al., 2008; Weiser, 1999; Strang & Linnhoff-Popien, 2004). Some approaches are for instance these in (Cai, Hu, Lü, & Cao, 2009; Chen, Finin, & Joshi, 2005; Kwon & Kim, 2009; Lee, Seo, & Rhee, 2008; Yang, Zhang, & Chen, 2008). It is worth to note that most of these works borrow technologies from the Semantic Web (Berners-Lee, Hendler, & Lassila, 2001).
Advances on QoS-aware web service selection and composition with nature-inspired computing
2019, CAAI Transactions on Intelligence TechnologyBuilding an open cloud virtual dataspace model for materials scientific data
2019, Intelligent Automation and Soft ComputingA novel patient friendly IT enabled framework for selection of desired healthcare provider
2019, International Journal of Medical Engineering and InformaticsA study on user-oriented and intelligent service design in sustainable computing: A case of shipbuilding industry safety
2017, Sustainability (Switzerland)