The project deals with the design of an intelligent sensor network for protecting premises from chemical, biological and intruder attacks. This project gives a logical level design along with the architectures at various levels of hierarchy. The use of object technology is proliferating in the development of software, and in order to build robust and maintainable complex systems, mastering object-oriented (O-O) analysis and design is essential.
The main goal of this study is to report on the experience of applying object-oriented modeling, analysis and design methodology to a real-world complex system represented by an intelligent sensor network for a building. UML has been used to model the software and automation infrastructure, which handles the interactions among processing elements in a modern building. A set of system design requirements are developed that cover the hardware design of the nodes, the design of the sensor network, and the capabilities for remote data access and management.
A formal model is proposed for the architecture, and the behavior diagrams explain the dynamic nature of the system. The static and dynamic diagrams together validate and verify the system. Agent UML is discussed to model evacuation of a room.
This project discusses some extensions to UML for agent-based modeling where the agents follow a purely reactive and proactive approach. In this work, agent-based architectures and behavior diagrams are proposed as a method to envision security in buildings. Extensions are provided to support a multi room scenario. Sensor fusion is used to provide a robust functionality and reducing the events of false alarms occurring in the system.
Linear programming techniques are used to solve for the minimal point in the cost vs. performance trade off curve for the sensor network as well as for the access system proposed. The tradeoff explores the relation of variables and suggests an operating point satisfying all constraints and without violating any requirement. Solver, an Excel add-in has been used to run the linear optimization.
Source: University of Maryland
Author: Varangaonkar, Rajeshree Vijay