Introduction to Multisensor Data Fusion

Course Description

Understanding the concepts, techniques, and issues surrounding the fusion of information from multiple sensors and sources of data. IST 885 Introduction to Multisensor Data Fusion (3)Rapid advances in nano and micro-scale sensors, ubiquitous wide-band wireless communications and improvements in computing provide the opportunity to collect and disseminate huge amounts of data and information from sensors, humans acting as observers, and emerging data available on the web. Applications for this data are widespread and include areas such as geospatial intelligence, emergency management, environmental monitoring, epidemiology, and others. This course introduces methods and process models for fusion of the information from diverse sources to achieve inferences that cannot be obtained by using a single source or sensor. Course Objectives: IST 885 provides an introduction to multisensor information fusion. Multisensor information fusion seeks to combine information from multiple sensors and sources to achieve inferences that are not feasible from a single sensor or source. the proliferation of micro and nano-scale sensors, wireless communication, and ubiquitous computing enables the assembly of information from sensors, models, and human input for a wide variety of applications such as environmental monitoring, crisis management, medical diagnosis, monitoring and control of manufacturing processes. Techniques for fusing multisensor and multi-source information are drawn from a variety of disciplines including statistics, data mining, artificial intelligence, estimation and control theory, pattern recognition, and signal and image processing. While this course is non-mathemathical it will help students understand the concepts, techniques and issues associated with developing and using multisensor data fusion systems. At the end of this course students should be able to: * Explain different models of multisensor data fusion and describe the advantages and limitations of data fusion * Explain the five levels of data fusion in the Joint Directors of Laboratories (JDL) data fusion process model * Assess and characterize a sample information fusion application * Identify various techniques used in multisensor data fusion and indicate the applicability and limitations of the techniques for a selected application * Design a data fusion system including specifying the required functions, applicable techniques, selection/assessment of sensors and information sources, and design of a sample user interface * Discuss current technology trends that affect the implementation of a fusion system. Student activities: The course consists of ten lessons and one capstone group project that will span either the 15-week semester or the combined 12-week summer session. Each lesson will require approximately 8 hours of student activity. Student activities will include reading lesson text, online quizzes, and discussions about the way in which multisensor information fusion is applied to selected applications such as geospatial intelligence, environmental monitoring, monitoring of complex systems, crisis management or related areas.