Predicting and Modeling Biological functions in Body Area Network

Recent advances in Wireless Body Area Sensor Network (WBASN) technology has become a leading approach for several promising applications in the medical field. WBASN is the network is built with different kinds of physiological sensors which measures different vital parameters on the human body for the purpose of monitoring the patients. Prediction and modeling are very two important issues which are required to address while building a WBASN. To enhance the long-term critical health monitoring, a robust predictive approach should be incorporated in every WBASN system and it leads to saving computation time and increasing the energy. In this paper, we describe the use of polynomial regression for predicting and modeling biological functions. We also describe how effective different orders of polynomials can be. There are four functions that we use for this purpose: blood pressure, scalp EEG signals, the walking gait of people with neurodegenerative disorders, and lastly motor movement signals. We have also used two different degrees of polynomial functions to determine the predictive value.