Research And Implementation Of Service-Oriented Spatio-Temporal Big Data Statistical Analysis Platform

With the popularization of location sensing and object monitoring technology and equipment,a large amount of spatio-temporal data has been accumulated in the fields of intelligent transportation and smart city.Spatio-temporal data is high-dimensional data generated by specific objects with both temporal and spatial attributes.As it is generated continuously along with the life cycle of objects,the amount of it is huge,so it is called spatio-temporal big data.Because it has both the technical characteristics of big data and diversified analysis needs,the traditional data analysis method cannot well meet the analysis and processing needs of space-time big data in many fields,especially in quickly meeting the statistical analysis needs of different businesses in different fields.Therefore,it has become an urgent problem to be solved how to conveniently select the dimension required by users from the multidimensional space-time big data and how to quickly conduct statistical analysis and dig the potential value behind the space-time big data.In view of this problem,this paper adopts the service-oriented software method to improve the development efficiency of the application of spatial and temporal big data statistical analysis,and mainly carries out the following four aspects of work:1.Aiming at the problem that the traditional methods cannot meet the needs of spatial-temporal big data statistical analysis,through the induction of the needs of spatial-temporal big data statistical analysis,two basic service models,namely data service and functional service model,are designed to support the statistical analysis of spatial-temporal big data.The data service is mainly encapsulated by multi-dimensional spatio-temporal data,which mainly provides the data source of analysis operation for the functional service.The functional service is summarized through the processing functions of the three links of the pre-processing,statistical calculation and result visualization in the process of statistical analysis.Two kinds ofservice models can provide guidance and normalization support for the servitization of spatio-temporal big data.2.Aiming at the low reusability of the above two kinds of spatial-temporal big data service implementations in the big data environment,a configurable service interface based on the big data environment is designed.The general data access service implementation framework and data query service interface are implemented.Typical spatial and temporal data preprocessing function service sets and configurable descriptive statistical analysis function service templates in functional services can improve the reuse ability of services and achieve the effect of independent customization and configuration of services.3.In view of users' diversified demands for statistical analysis of spatial-temporal big data,an application construction method of spatial-temporal big data statistical analysis based on service combination is designed.Methods combining the characteristics of data service and functional service,a spatial-temporal big data statistical analysis application was constructed by combining the forms of basic data service and functional service in turn.At the same time,the definition language of spatial and temporal big data service composition is given based on the way of extending BPMN.4.In view of the construction of spatio-temporal big data statistical analysis application,a spatio-temporal big data statistical analysis platform supporting the above methods was designed and implemented by using open source Flowable software and Hadoop big data environment.The platform mainly consists of three levels of functions,namely,spatial-temporal big data access management,spatial-temporal big data statistical application construction and application realization,which can provide fast construction and deployment execution support for spatial-temporal big data statistical analysis applications.At the same time,through the application of the platform in the statistical analysis of big data of expressway toll,the effect of the platform is verified.