Dealing with large schema sets in mobile SOS-based applications

Abstract

Although the adoption of OGC Web Services for server, desktop and web applications has been successful, its penetration in mobile devices has been slow. One of the main reasons is the performance problems associated with XML processing as it consumes a lot of memory and processing time, which are scarce resources in a mobile device. In this paper we propose an algorithm to generate efficient code for XML data binding for mobile SOS-based applications. The algorithm take advantage of the fact that individual implementations use only some portions of the standards' schemas, which allows the simplification of large XML schema sets in an application-specific manner by using a subset of XML instance files conforming to these schemas.Comment: 9 pages, 2 tables, 7 figure

Personalised code generation from large schema sets for geospatial mobile applications

“…After applying the schema simplification algorithm it was determined than only 327 components and 423 relationships are needed to process the input instances [5], representing only 9.81% and 9.16% respectively of the total number of component and relationships. Special importance among these components have complex types, which are used as primary concepts to generate code, that were reduced from 846 to 112 types (13.23%).…”

“…The mobile client application is referred to as Air quality mobile client in the remainder of this article. We have chosen SOS because of our previous experience with this specification in the server and client side [5,14,40], and also because the SOS schemas are among the most complex schemas of the OGC specifications with more than 700 types distributed over more than 80 schema files. These figures undoubtedly represent a real challenge for testing our approach.…”

“…While geospatial applications such as navigation systems, web mapping services and location based services are more common everyday on mobile devices, mobile applications that interact with OWS services are still a few. In our opinion, though, the complexity of some data encodings makes it difficult and time-consuming to create reliable and efficient OGC-compliant mobile applications [5]. For example, OGC services may serve huge amount of observational data packed in complex data structures in XML format, which is reputed to be extremely verbose in certain scenarios [6].…”

“…Built on our previous works [5] and [13], in this article we present an approach called Instance-based XML Data Binding.This approach is based on the observation that the existence of large schemas does not imply that every single XML element and type definition is required by client applications [14]. That is, a client application is built according to the real requirements rather than supporting the full range of data structures offered by a given service specification, being most of them particularly unnecessary in that client application.…”

Measuring complexity in OGC web services XML schemas: pragmatic use and solutions

“…On the other hand, the amount of increasingly structured data exchanged between clients and servers has reached levels that may represent a serious challenge when building a real system. For example, (Tamayo et al 2011c) present the obstacles to produce XML processing code for a mobile client for the Sensor Observation Service (SOS) specification (OGC 2007g) using existing XML data binding technologies. As a consequence of the large size of the SOS schemas, the code generated by these tools exceeds the hardware limitations of mobile devices.…”