University of Twente Student Theses

Login

Architectural patterns for developing geospatial web and mobile applications

Yaghoubkhani Ghiyasvand, Morteza (2021) Architectural patterns for developing geospatial web and mobile applications.

Link to full-text:https://ezproxy2.utwente.nl/login?url=https://library.itc.utwente.nl/login/2021/msc/gfm/yaghoubkhanighiyasvand.pdf
(only accessible for UT students and staff)
Abstract:The rise of the information communication technology (ICT) and infrastructure, the recent technological advancements in computational power, and the introduction of new tools such as smartphones, tablets, and laptops, are giving service providers and software developers a unique opportunity to provide more and more functionalities and services through the web. Thanks to this opportunity, like others, GI scientists and professionals use the web as a new infrastructure to provide GIS services. Today almost all desktop GIS functionalities could be provided on the web using geospatial web applications. A geospatial application uses web technologies to obtain data and information, process this data, and disseminate it. It also uses the web for communicating between these three main components. Compared to desktop GIS, geospatial web applications have several advantages, and the most important one is that their development process is cheaper and faster. There is no need for updating software and data in geospatial web applications. These applications have higher levels of accessibility to users, and they are better suited to data acquisition and dissemination on the network. However, there are some issues in developing geospatial web applications that cause some disadvantages. Along with technological advancement, users expect more functionalities from geospatial applications. Users are always looking for more advanced analysis on massive geospatial data (with various data types from different resources) in less time. That is why geospatial web applications should be able to access and handle massive amounts of spatial and non-spatial data with different formats from multiple sources. Another major issue in geospatial web application development is rapid changes in web development technologies and platforms, making it hard for the applications to keep up with the pace. Since most of these applications are based on a particular technology, they cannot be updated for the new technologies and have to be developed from scratch. Also, issues regarding application security and spatial privacy are among the most significant geospatial web application development challenges. Several web application development methodologies such as Model-Driven Development (MDD) or Data-Driven Development could help developers deal with these issues. This research proposes the Pattern Based Model Driven Architecture (PBMDA) for geospatial web application development. The PBMDA approach is based on the Model Driven Architecture (MDA) and tries to integrate software design patterns into the MDA process. MDA is based on the MDD notation and emphasizes the use of models with different abstraction levels in all phases of the web development process, and software design patterns are standard solutions for recurring problems in software development. Using the PBMDA approach, we can ensure flexibility, interoperability, better communication, and inclusion in the web development process. Based on the PBMDA approach, we first developed a Domain Specific Language (DSL) for a particular geospatial web application. Then, considering the technical specifications related to the application’s implementation and preferred platforms (implementing using a JavaScript stack, e.g., MongoDB, ExpressJS, ReactJS, and NodeJS), we created a Platform Specific (PSM) metamodel. We generated a PIM-PSM transformation function using pattern-based model transformation (PBMT) with PIM and PSM metamodels. Next, the Computation Independent Model (CIM) has been developed using the application logic. Using created CIM and generated DSL (PIM metamodel), we have developed the PIM. Finally, using the PIM and transformation function, we have generated the PSM. The application code generated automatically from the PSM.
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:https://purl.utwente.nl/essays/88971
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page