无线网络中QoS感知的多粒度服务组合研究

Characters)：In recent years, the remarkable advance of wireless networks and communications bring the requirements on applications in wireless networks into an extend area. The integration of different kinds of wireless mobile networks is also emerging. Mobile networks and devices supporting multiple air interfaces are regarded as one of the most important features for future mobile systems. The success of these mobile systems lies in their ability to provide users with cost-effective services that have the potential to run anywhere,anytime and on any device without(or with little) user attention. Recent popularity of smart phones and pads also adds more demand on providing rich application services on mobile handsets. One important and effective enabling technique is service composition. Service composition is treated as a promising technology in resource reusing, integration, sharing and providing value-added application services in an agile way. It is naturally for the research hotspot of service composition to move into the wireless networks from traditional wired networks. When conducting service composition, certain constraints or Quality of Service(QoS) have to be considered, namely, QoS-aware service composition. This usually refers to the problem of composing a set of appropriate services into a richer service that follows application logics while satisfying certain QoS requirements. This project is to research the QoS-aware service composition in wireless networks. The wireless network environment is categorized in two classes which are semi-static environment and dynamic one. Service composition oriented task partition is to be studied to increase the effficency of service composition planning with energy saving.Research on parallel branch service composition is to be carried out to avoid the interference of wireless bottleneck area on service composition. Through the research on hierachical multi-grain service composition model and algorithms, multi-criterias of QoS, resource restraints and application requirements are to be integrated in one frame to optimize the service composition. Multi-stages multi-steps service composition is to be researched in this project to solve the problem caused by dynamic and frequent variance of QoS and network environments.The simulation and prototype will be put forward to evaluate and demonstrate the research results.

随着无线网络通信技术的发展，对无线网络中应用服务的需求越来越广泛。服务组合作为一种资源重用、整合与共享以及快速提供增值应用服务的技术，其研究热点从传统的有线网络领域向无线网络领域发展是一种自然趋势。除了满足功能性需求，服务组合还要为用户和具体应用提供服务质量(QoS)保证。本项目研究无线网络环境中QoS 感知的服务组合问题。拟将无线网络环境分为半静态和动态两种类别。将对半静态无线网络中组合服务的划分问题以及并行分支服务组合算法进行研究，从而在移动设备能量受限的情况下提高服务组合的性能，避免无线瓶颈区域对服务组合性能的干扰，采用分层多粒度服务组合模型与算法，综合考虑无线网络中服务的各种QoS 标准，资源限制和应用需求，以得到优化的服务组合结果。通过研究分阶段多步的服务组合模型，为动态无线网络中服务组合面临的网络环境和QoS 动态多变的问题提供解决思路。研究结果将采用仿真和原型系统进行验证。

为了确保以及提高无线虚拟网络中数据服务的服务质量，减少成本，从而最大化用户和虚拟运营商的效益，我们提出了一个基于无线虚拟网络的服务计算框架。在这个框架的基础上，分别考虑云无线接入网络环境下的数据服务速率优化问题，无线虚拟网络中的数据成本优化问题，移动自组织网络中的拓扑结构优化问题以及软件定义的网络的性能评估问题。模拟仿真结果显示，我们提出的模型和算法具有良好的性能。相关的研究成果发表在SCI,EI检索的国际期刊和国际会议上。此外仍有一些后续工作在投稿中以及准备中。这些研究的成果为将来进一步研究大数据环境下服务计算中的服务质量优化和服务性能评估问题做好了准备。