Research On Fusion System Framework And Algorithm Of Multi-vital Signs In Body Area Networks

With the maturity of information technology and the popularization of wearable devices,the technology of body area networks(BAN)has been developing rapidly,which makes it easy to monitor vital signs of human body anytime and anywhere.With the improvement of the quality of life in recent years,people pay more attention to their own health in daily life.BAN adapts to the needs of social development and has a broad application prospect.BAN can provide a feasible remote medical monitoring program for the empty nest elderly,and can provide appropriate fitness recommendations through real-time monitoring of people's vital signs,and can also conveniently monitor the athletes' movement status without limitation of space.However,there is no unified standard in data fusion field which is the key technology of BAN.Further research is needed to judge the user's physical status more accurately and efficiently based on various physical signs collected by sensors.The research contents of this thesis are summarized as follows:1.On the basis of traditional data fusion algorithm,an adaptive multi-vital signs fusion system framework based on subjective and objective weight is proposed,and its working process is explained.Because different physical signs have different effects on the decision results,the framework effectively combines the subjective weight which represents expert opinions and the objective weight that is derived from data reasoning,and adjusts the contribution rate of the two weights through user feedback to get more accurate decision results.2.There are many kinds of sensors which have different accuracy in multi-vital signs fusion system.The current data fusion algorithms have low accuracy and poor applicability when dealing with the data of conflict and uncertainty.Aiming at the above problems,a multi-vital signs fusion algorithm based on evidence distance and membership degree is proposed in this thesis.The algorithm uses evidence distance and membership degree to measure and quantify the degree of conflict between different sensors,which can effectively reduce the impact of abnormal data in the fusion process.It is verified by a classic example,which shows that the algorithm can effectively improve the reliability of decision results.3.The multi-vital signs fusion system inevitably contains a large amount of historical data and real-time data.In order to make full use of the effective information of the two kinds of data,a mixed support degree based multi-vital signs fusion algorithm is proposed in this thesis.The historical support is calculated by historical data based on information gain and the real-time support proposed is calculated by real-time data based on evidence distance.Then the combination rule of Dempster-Shafer(D-S)evidence theory is improved to effectively solve the data conflict problem.In addition,in order to improve the adaptability of the data fusion algorithm,the feedback of users is introduced to dynamically adjust the fusion process.The final simulation results show that the algorithm proposed in this thesis can effectively improve the accuracy and reliability of the decision.The above research provides a new effective mode for the design of multi-vital signs fusion system framework in BAN,and provides a more precise and intelligent algorithm for multi-vital signs fusion technology,which lays a solid foundation for the wide application of BAN.