Research On Multi-sensor Information Fusion And Its Application

In recent years, with the development of the sensor network, the number of sensors is increasing,and the distribution of the sensor is more widely, so information sharing has become an irresistibletrend. The basic information which obtains by organizing sensor information effectively is tomaximize the intelligent information service. Meanwhile, multi-sensor information fusion can getthe information which single sensor unable to obtain. This is an important topic in the field ofinformation service, so that’s where this thesis coming from.This thesis introduces the related theory firstly, including the principle of multiple sensorinformation fusion and technology of current information fusion. Then, it focuses on the research onarchitecture and algorithm of multi-sensor information fusion. This thesis describes a newinformation fusion model based on the existing function model of information fusion, whichcombines the advantages of existing model. Not only taking the level of information fusion intoaccount, but also can it meet the functions of information fusion. So we can focus on the design ofthe method of information fusion. As the level of information fusion, the data layer algorithm,characteristic layer algorithm and decision-making algorithm is discussed respectively. It adoptsimproved pre-treatment method of the D-S Evidence Theory based on current domestic andinternational research on the evidence conflicts for resolving the problem of evidence conflicts.To validate the effective of the improved information fusion algorithms and reasonableness ofinformation fusion model, an evaluation experiment platform of the quality of householdenvironment is developed. It implements actuator control, as well, acquisition, storage and sharedwithin the LAN of sensor information.