Key Technology Research Of Device Management In Ubiquitous Stub Enviornment

As the development of network towards high speed and ubiquity, a large number of heterogeneous devices are distributed in the edge of network and surrounding of users, forming the ubiquitous stub environments which provide ubiquitous services for users. The ubiquitous stub environment brings a great challenge for traditional device management, as it is not only required to cover the requirements of traditional device management but also essential to manage the new requirements brought about by the ubiquitous stub environments, such as device coordination mechanism etc. Device Management in ubiquitous stub environments extends the traditional network management area, enriches the research content of network management and plays an important role in promoting the development of network management technology.In ubiquitous stub environments, multiple devices cooperate among themselves to establish the intelligent environment for developing the distributed ubiquitous services and supporting high-quality ubiquitous services for users. How to achieve and manage the collaborative work among multi-devices to satisfy the users’requirements for ubiquitous services becomes an essential issue of Device Management in ubiquitous stub environments. Meanwhile, with the growing enrichment of device capability and its supportive service type, faults and hidden problems tend to be more likely to occur, especially for mobile devices. How to manage bearer services on mobile devices and their own faults to guarantee the normal use of users in daily life comes to be another key problem of Device Management in ubiquitous stub environments. Therefore, based on the analysis of current research results related to Device Cooperation Management and Device Fault Management in ubiquitous stub environments, this paper combines existing related Network Management technology to develop the deep research on those problems. In the aspect of Device Cooperation Management, a service-oriented device cooperation management mechanism is proposed, mainly including the mapping model between service and network, and critical device cooperation algorithms which can be applied to both scenes of single service and multiple services. In the other aspect, a device fault management method oriented to the service availability is created in terms of bearer services on mobile devices and their own fault management. In this paper, the main research contents and innovations are as follows:(1)In ubiquitous stub environments, it lacks network and service relationship model oriented to the service quality. So a service-oriented multi-devices cooperation mechanism is proposed. Firstly, a ubiquitous service logic model and a ubiquitous stub network model composed of devices are proposed in this mechanism. Then a QoS-oriented network-service mapping model is constructed in the form of sub-graph isomorphism. Next, cooperation discovery, cooperation selection and cooperation maintenance are proposed to describe multi-device cooperation process based on the problem model. At last, multi-devices cooperation algorithm is proposed to ensure high quality ubiquitous service in ubiquitous environments.(2) Based on the Service-oriented Multi-devices Cooperation Mechanism, we further consider the problem that during the execution process of single ubiquitous service, network topology changes with devices’movement, which increases the disruption times of ubiquitous service and the jitter amplitude of service benefits, and a service-oriented multi-devices cooperation construction algorithm is proposed. Firstly, we introduce the time series to establish a dynamic ubiquitous environment model, and make normalization of service quality indicators. Then a single-service-oriented multi-devices cooperation model with double objective optimization problem is formulated to describe the problem. Finally, we introduce the backup device cooperation set strategy with periodic update, based on which, a single-service-oriented multi-device cooperation construction algorithm is established to avoid ubiquitous service disruptions and service quality jitter caused by the dynamic nature of ubiquitous environments.(3) In ubiquitous stub environments, a multi-service-oriented device composition algorithm is proposed for the circumstance when multiple users request for services simultaneously. Firstly, a weight calculating method based on TOPSIS multi-attribute decision-making and comprehensive variance maximization is introduced to model the problem. And it is transformed to typical MMKP by means of dimension reduction. Then, WS-HEU was imported to select device cooperation sets to perform the corresponding multi-services. In this way, it avoids ensuring one specific user’s service quality at the expense of others’ experience and increases the utility benefit of ubiquitous stub resources, hence laying the foundation to support autonomous services in ubiquiouts stub network in the future network.(4) In order to detect, diagnose and exclude potential or occurred faults on mobile devices, a user-oriented detection mechanism is proposed. In the mechanism, a User-Behavior Analysis model based on fuzzy math theory is firstly proposed to weigh the dependence degree of each service. Secondly, a User-Behavior Clustering Analysis model is introduced to divide mobile users into several clusters. Thirdly, the paper integrates the dependence degree of each service, the network status and failure rate of service into building a priority detection-service set selection model, which could select an optimized set for Pre-Detection of the service faults on mobile devices.(5)For another device fault management problem, we firstly propose the architecture of mobile device fault management, and describe the challenges and requirements in details. And then the managed resources are modeled for a proper management pattern, based on which the concept of Case-based Reasoning for discrete fault events is imported. At last, the prototype system based on OMA DM is designed.