The Research Of Structure And Algorithm For The Moving Evaluating System Based On Physiological Information Fusion

In recent years, with the development of information diversification, microelectronic technique and multi-sensor network, research on information fusion has been developed rapidly as an effective method to process information synchronously. The information fusion has been applied to many areas, such as automatic target recognition, battlefield surveillance, automatic flight vehicle navigation, robot, remote sensing, medical diagnosis, picture processing, pattern recognition and complex industry process control and so on. The application in physiology field is still in the junior stage. The study mainly focused on qualitative medical diagnoses, and the dynamical physiology fusion is deled with comparative shortly. With the development of IT, dynamical physiology fusion surely is able to get more and more broad attention, especially in the sports health, military training as well as the outer space imitates.By fusing much body physiology information, we can not only gain the estimation about human body situation, but also valuate the moving process or training effects by fusing and monitoring the body physiology information under special appointed state. Since the physiology information signal is weak, noise is strong, and non-stationary, the physiology information fusion have certain particularity. From actual use, and by combining the fusion theory, a system model was proposed in this paper based on physiology information fusion. The model was used to evaluate the moving effect, and was explained in structure and function detailed in this paper. More algorithms have been analyzed, and the nuclear algorithm was developed based on distributed Kalman filtering and fuzzy BP neural networks. Through the moving platform provided by the subsidiary hospital of Hebei University, the data about blood pressure, heart rate, and pulse and oxygen saturation of blood in the moving were collected. These data was used in the GUI to carry out an example. Simulation result showed that the algorithm is validity and feasibility.