MECA: Mobile Edge Capture and Analysis Middleware for Real Time Decision Making

In real time decision making scenarios (e.g., a coalition environment), relevant and accurate information must be made available to decision makers in a timely manner. However, decision makers face a dual problem of both information overloading and starvation: the overwhelming majority of information from various sources may not be relevant to the investigation goal, while a significant gap exists in useful and pertinent information. To address such problems, we propose and develop MECA, a common middleware infrastructure for data collection and analysis from various data sources, especially modern mobile devices. MECA provides a high level abstraction of a phenomenon specification such that applications can accurately express their information needs in a declarative fashion. Thus only data relevant to the investigation are gathered. MECA supports primitive processing within the infrastructure to transform raw data to information at semantic levels much closer to decision making. It coordinates the data collection and transformation activities, such that data and information can be shared across applications without duplicate efforts. We showcase the capabilities of MECA through disaster management and military applications. Finally we discuss how to expand it to further automate the decision making process. keywords: data collection, decision making, real time