| With the development of the observation technology and the great progress made in acquisition of spatial data, the produce of the great volume data, with the help of high speed network and the complicated computing technology result in one another heterogenous data producer and data comsumer. But how to integrate those data from different data providers and applications fields, and make the data into knowledge and share them between all data providers and data users has become a hot topic. With the rise of "digital earth" and spatial data infrastructure (SDI) construction, providing technical support and implementation to achieve spatial information and services sharing and interoperation under distributed network environment, in particular, the construction of spatial data infrastructure, making the earth and geographic-related spatial information can be expressed,decrypted and queried on the network. Spatial data infrastructure defined geospatial data framework and a variety of spatial data exchange standard. In addition, in order to achieve a variety of web services sharing and interoperability, The Open Geospatial Information Committee(OGC)Services Initiate projects (OWS)and the ISO TC/211 and other international organizations define various types of annotation interoperability specifications, such as Web Map Service(WMS) Implementation Specification, Web Feature Service(WFS) Implementation Specification, Web Coverage Service(WCS) Implementation Specification. However,as geo-spatial information services, types of diversity among the different types of services vary greatly, and the same kinds of services, due to the different versions base on the information model, resulting in a variety of different versions of the service, different types of services. Different versions of the same types of services operating between OGC specifications also can not be achieved under the action of a direct interoperabilitity, at present, different versions of the same types of access between services, OGC implementation of the specification to provide a version of the so-called consultative mechanism. This mechanism has two major drawbacks:1. Can not satisfy the user an accurate version of the service reques; 2. Reliability is poor, there is a potential risk of failure of consultations. Therefore, there are now multiple versions for heterogeneous network sharing and interoperability between service problems. According to different types and different versions of the same type the difference between the network service model and associated, connection with the network service model matching and document information extraction and conversion of the relevant theories and methods, to carry out open-geospatial information network services, unified multi-version access methods of research has important theoretical and practical significance.This geospatial information in multiple versions of different versions of network services model to study, according to different versions of the pattern of similarities and differences between files from different versions of web services pattern matching method and different versions of the service request response to the outcome document of information extraction and conversion technologies start to design a more uniform access to middleware versions of network services, through different versions of files on the dynamic matching model and the results of different versions of the request for information extraction and conversion to achieve the different versions of the exact response to service requests, the main results of research and innovation as follow:1) Based on the open-geospatial web service model characteristics of different versions of schema files, make a comprehensive analysis of the current general schema-matching method and proposed a grammar schema matching method based on schema fragments. Focusing on the schema dynamic loading and presentation, the partition of the schema files, the recognition of schema fragment-pairs, and the fragment matching techniques. And use the different versions web service schemas deployed in Geoserver server to do the match experimental. The experimental results show that the transformation of a big schema matching problem into smaller fragment-based matching problem, due to the reduction of matching size and complexity, and achieved not noly a better matching quality, but also improves the matching efficency greatly. Compares to generic schema matching methods, such as COMA, the matching quality of fragment-based matching method improves about 5%, the efficiency of schema matching improves more than 30%.2) To analysis the semantic heterogeneity problems brought by the difference of different version web service model, according to the problem caused by name difference, for instance, apply a gramma based schema matching method can not find those mappings with name difference. Therefore, this paper presents a semantic matching method based on node similarity, study the status of semantic matching and concepts about semantic matching. In two different versions of the web service model diagram, for example, detailed analysis of the establishment of label concept and node concept in semantic matching approach, focusing on the computation of relationship between label concept, building formulation of relationship between node concept and transform the computation of node concept relationship into propositional formulas validation. And apply the semantic matching method to do the match works based on gramma schema matching tasks, the experimental results show that the application of semantic schema matching method greatly improves the recall rate of schema matching.3) Study technology about information extraction and the XML document conversion-related, study on how to use schema matching results to generate rules file for different document information extraction and conversion, to archieve rules file generation from a variety of different versions of web service model match results, and finally apply the rules file to complete different document information extraction and conversion by using XSLT document conversion techonology, and analysis the information extraction efficiency in end.4) Application of service-oriented architecture (SOA), the schema matching and information extraction features packed into a middleware for unified access to multiple versions of web services, and the middleware applied to the Antarctic Spatial Data Infrastructure for the unified acdess to SCAR KGIS and SCAR ADD multiple versions of web services.This paper explore current problems in access multi-version open geospatial web service and put forward an effective solution, and through different web server at home and abroad to provide a number of different versions web services schema files for experimentation of schema matching and response to a request document information extraction and conversion to verify the correctness of the related theory and the feasibility of the methods.
|
PDF Full Text Request