Study On Pulse Data Fusion In Body

Body sensor network as the representative of new technology can deploy biosensors on the body and transmit the physiological data to the monitoring equipments in wireless way. By analyzing these physiological data, it can play a role in the prevention of disease and a method is provided to "cure not ill". Cardiovascular disease has become the leading killer to human health because of its sudden. The World Health Organization states:there are not yet effective vaccines to treat cardiovascular disease except prevention. Many researches have shown that pulse wave’s propagation characteristic is closely related to many physiological parameters of cardiovascular system. Therefore, the feature extraction and fusion of pulse wave information in the network is done based on body sensor network, providing online monitoring and prevention for cardiovascular disease.A pulse-feature data fusion algorithm is proposed based on the systematical analysis of pretreatment, feature extraction and data fusion of pulse signal. Combining the algorithm with LEACH routing protocol, a data fusion routing protocol based on D-S evidence theory is proposed. The specific research works are as follows.The wavelet threshold denoising algorithm is used to complete the pretreatment. From two angles of the signal-to-noise ratio and the root mean square error, many experiments have been done to calculate the optimal wavelet base and the optimal decomposition layer for the wavelet threshold denoising algorithm. The representative pulse wave is calculated by theoretical analysis, and then finds its peak, valley and average values to calculate the eigenvalue of K value. The mode energy ratio is calculated by empirical mode decomposition, and then completes feature extraction process.The calculation methods of basic probability assignment and correlation degree are introduced on the basic of D-S evidence theory, and the fusion process of K value and mode energy ratio are compeleted by the D-S evidence theory. Combined with three aspects of pulse signal pretreatment, feature extraction and D-S evidence theory, a pulse feature fusion algorithm based on D-S evidence theory is proposed, and the feasibility of the algorithm is verified by experimental data analysis. On the basis of LEACH procotol, a data fusion routing protocol based on D-S evidence theory is proposed to reduce the amount of data transmission in LEACH protocol.Through analysis of the experimental data, results of the pulse signal pretreatment, feature extraction, data fusion can reflect the real information of the tested object. The simulation results show that compared with LEACH protocol, the data fusion routing protocol based on D-S evidence theory reduce the amount of network data transmission, save nodes energy and prolong the network life cycle.