Computing aspects of monitoring walking disorder using body sensor network and neural network

Here, it is proposed to monitor walking disorder of any patient with the help of wireless three dimensional (3D) accelerometer based body sensors and its networks. We gather Ground Truth Data from the sensors, filter it, boost up it when required, then collect some important features and compare with the features of run time data using different algorithms developed by us. Where, for matching the run time features, we use supervised learning method of back propagation neural network. After gathering data, a prototype model of computation is developed which may be used in any motion disorder of any subjects like: patients, athletes, pilots and astronauts. The contribution of this paper is focused on to develop a model of computational processes required to monitor activity recognition system. The computing model developed is validated working over different walking motion disorders of different subjects and then discussed how this model can be applied in an organization to provide internet based online patient´s information services with the help of Feature Server and Local Server connected by wireless radio link of Personal Computing Devices like mobile phone, PDA, laptop etc.