New Criteria for Output Strict and Input Strict Passivity for Interfered Digital Filters for Biomedical Applications

The biomedical signals are processed with various digital filtering and pattern recognition techniques which are used to collect and analyze the data to capture the important features of a given biomedical system. In a medical apparatus, a set of these techniques essentially work in real time. It contains the vital communications hardware and an advanced level software interface which enables the user to work in real time mainly for biomedical applications where processing speed, stability and accuracy is very important. This paper addresses the stability issues of digital filters in the presence of external disturbance to the real time biomedical computing system. Specifically, new output strict and input strict passivity criteria for fixed-point state-space digital filters with saturation arithmetic in the presence of external interference for biomedical applications are established. The proposed linear matrix inequality-based criteria confirm the asymptotic stability of the digital filter in the absence of external interference. The proposed approach may produce larger stability region in the parameter-space which will be useful for the processing of biomedical signals and images. Two numerical examples are given to illustrate the usefulness of the presented criteria.