Study On Information Fusion Of Intelligent Garment-Oriened Body Area Sensor Network

To embed electronic information technology into clothing worn by human beings at all times,on one hand will bring to cloth more functions,in addition to keeping warm, beauty,on the other hand will also change the look of the original information system in an unprecedented manner.Intelligent garment,taking into account the integration of information technology clothing as the fundamental starting point,taking advantage of materials,textile,apparel and other fields of advanced technology,implant detection, treatment,storage,communications,and I/O micro-electronic components into our clothing flexiblily and without affecting the comfort of wearing apparel for the purpose of real-time signal acquisition,processing,storage and transmission of data and various applications,to provide users with intelligent analysis of decision support and feedback control.Wearable computing,sensor networks and information fusion provide for the intelligent garment with the computing platform,the communication platform and decision-making plateform respectively.Intelligent garment,which closely around human body,has been forming a unique area of research and of significance in theoretical research andd application prospects.First,the thesis systematically expoundes smart clothing and related areas as research results.After the introduction of the current study status quo of hardware platform, software architecture,network transmission and information processing of the smart clothing,we explore main methods of the physiological information processing and multi-source information fusion.Finally,we discuss the design philosophy,system architecture and application area of intelligent garment and make some prospects on the further work of the study issues in this field.Second,from the complexity the intelligent garment system and the need to adapt to the changing application and the changing human body,we provide the characteristics of the actual demand of intelligent garment oriented physical monitoring system and the methodology of building such a system.We design the layered architecture of intelligent garment.The design philosophy,laws,and solutions are also given out.Considering the basic application of the intelligent garment:human physiological monitoring,we provid a design of hardware and of signal processing modules.Then,based on genetic algorithm we use multi-objective optimization of task scheduling to meet systems performance requirements,such as low-power,real-time and balanced computing.Third,the intelligent garment needs a flexible low-power communications mechanism to effectively transmit sensor data and control information.We use low-power IEEE 802.15.4 WPAN technology embedded in clothing to provide the framework of the communications platform.According to the needs of the physical sensor network data transmission of the health care monitoring application,based on the IEEE 802.15.4,we design body area wireless sensor networks embedded in clothing, determining the deployment of its nodes,network band,topology,media access patterns, and other related network communication mechanism.The simulation has proved its viability.Then,we explore comprehensive analysis methods and a decision-making system, which uses information of various human physiological signals,activities signals and environmental signal acquired in real-time by acquisition system embedded in clothing. It is very important and difficult to determine whether the person who is undertaking physical exercise is safe.A unique approach based on both analytic hierarchy process (AHP) and fuzzy comprehensive evaluation(FCE) is presented to evaluate exercise risk. First,a new evaluation model combining the above two methods,AHP-FCE,is established by applying the fuzzy mathematics into hierarchical assessment.Then, various physical exercises relevant factors are analyzed and personalized factors are determined basing on grey correlation analysis.In addition,steps of the AHP-FCE based physical exercise evaluation are set up.Finally,an actual calculation example is used to verify the feasibility of the method.Finally,the chronic fatigue symdrone(CFS),usually called "sub-health" is a generally concerned human physical and mental health problems.Because of its complex forms and subjective vague characteristics,diagnostic technology of sub-health makes slow pace of development.We provide a comprehensive quantitative analysis method integrating physiological parameters based on our new intelligent garment. Physiological signals are sensed by using the human physiological sensors embedded in the information clothes,features are extracted and signals are processed through the distributed computing platform and the sub-health assessment model are established based on the physical information fusion.Principal component analysis is used to perform feature dimension reduction.Based on the physiological database,the algorithm is verified and the results are analyzed.Out work will not only advance the public health and medical response to CFS,but also server as a model for other chronic medically unexplained conditions such as post infectious fatigue,cancer-related fatigue.In this thesis,we study the information fusion of intelligent garment oriented body area sensor network,aiming at establishing low-power flexible human life signal perception and transmission network,at exploring for medical,sports and other physical and psychological analysis application-specific information fusion methods,and at providing a powerful theoretical framework and system prototype platform for the key functions of intelligent garment technology.Intelligent garment,as cross-hot of many technology fields,has an important significance of application.The research of body area sensor network provides a viable and flexible information perception sensing tools and platform for information on clothing.The wearable computing explored in this thesis is one of the core issues of ubiquitous computing research and is important content of mobile computing research.The intelligent garment based various application systems,such as health care,disease diagnosis,body posture awareness, physical exercise risk analysis and emotion recogonition,display model and effects of information fusion and intelligent decision-making of intelligent garment through a multi-angle.These works theoretically abstract and simplify the complex process of human physiology information processing and provide a new perspective and new ideas for the intelligent garment related application reseraches.