The quantity of data available to decision makers of various types is rapidly expanding beyond the pace of manual interpretation techniques (Hobbins, 1). Introducing a Service Oriented Architectures (SOA) based web service framework that exposes even more data without sufficient guidance will exacerbate the situation. Ontology's, data descriptions and discovery methods alone are not enough to create the end-to-end solutions promised by SOA technologies. Software architectural patterns in conjunction with broad data strategies are required to harness and employ vast quantities of content. This dissertation provides two software architectural patterns and an auto-fusion process that guide the development of a distributed, accountable and scalable SOA framework to support improved control and monitoring software. Although applicable to a wide range of software control system challenges, the dissertation will focus on a Maritime Domain Awareness (MDA) interoperability challenges. Using the U.S. Navy's MDA project as a case study, this dissertation will design, build and test a prototype automated data fusion framework employing the trickle-up and Command and Control Zone pattern that automates the discovery, pedigree assessment and ultimate fusion of dissimilar data types in a SOA web-service supported framework.
展开▼
