Application Research Of Multi-sensor Data Fusion In Human Health Monitoring

With the development of science and technology,people's living standards are constantly improving,and more and more people are sub-healthy and even unhealthy.At the same time,the demand for the medical industry is also increasing.Due to the limited medical resources available,it is impossible to conduct comprehensive and professional health monitoring for most people in the current society.In view of the above problems,this paper proposes to apply the multi-sensor data fusion algorithm to human health monitoring,collect human physiological parameters through wireless sensor network,and then use data fusion algorithm to analyze the physiological parameters,and finally get the health status of our human body.This article takes human physiological parameters as the research object.Firstly,for the existence of outliers in the raw data collected in the wireless sensor network,an outlier recognition and detection algorithm based on multi-signal data correlation is proposed.Using the correlation between multi-modal data,the abnormal point data is identified and detected.Compared with the traditional anomaly data detection algorithm,the algorithm has higher recognition rate and lower false positive rate.Secondly,As the basic algorithm of data fusion,DS evidence theory proposes a method combining data of support vector machine and DS evidence theory for data fusion and decision making.Aiming at the shortcomings of the kernel function and the penalty factor in the standard support vector machine classification process,it is difficult to determine the optimal value.The dynamic chaotic firefly algorithm is used to optimize the kernel function and the penalty factor.Experimental simulations show that this method improves the classification effect compared to the standard SVM algorithm.The optimized support vector machine is combined with the DS evidence theory to form a new data fusion model.Through experimental simulation,it is found that the algorithm has certain advantages in recognition and decision-making compared with the standard DS evidence theory and support vector machine.Finally,this paper constructs a human health monitoring software and hardware platform,and combines the collected human physiological parameters with the data fusion algorithm model in the fourth chapter of this paper,and analyzes the results accordingly.The above experiments show that human health monitoring based on data fusion algorithm is feasible.The assessment of big data health and health status is the trend of future development.The research of this topic can help us and medical workers to detect their status more effectively.It can be seen from the simulation experiment results that the model studied in this paper has certain advantages in comparison with traditional algorithms in the detection and identification of health status,and provides some ideas for future healthy development.