A novel technique for document filtering and delivery for a pervasive computing model using selective dissemination of information

Mobile communication is booming and access to the Internet from mobile devices has become essential for the society to have information accessible from anywhere, any time and any device. Additionally, the exploding volume and diversity of digital information provide challenging issues in large-scale wide area information systems and introduces the need of delivering the right information to the right people. This scenario motivates this thesis research work. In particular, in view of the above statement, this thesis research work propounds the mixture of technologies like pervasive computing, mobile communication systems, network design architectures, selective dissemination of information, with platform, location and device independent information interchange specification like XML (Extensible Markup Language), and web services. In this work, the publish/subscribe system for a pervasive computing model is investigated for profile driven information filtering and retrieval. Here, user profiles are modeled by the XPATH profile modeling language. The digital information to be searched and retrieved is modeled by XML documents in NITF (News Industry Text Format) standard. In particular, an XML Filter is proposed and developed to receive XML documents and XPATH user profiles. As we show the information is filtered in accordance with user profiles based on XML-based-selective-dissemination-of-information. The filtered output is delivered to mobile clients using mobile web services, web services description language and simple object access protocol. The performance, information retrieval time, scalability and figure of merit of the proposed filter engine are investigated and explored with reference to a delivery mechanism adapted to the mobile devices. Finally, the performance of the proposed filter engine is modeled to investigate and validate the experimental results obtained using the proposed architecture for selective document filtering and delivery in a pervasive computing environment. The experimental results presented show the potential and value of the proposed method.