Optimisation of the Transient response of a Digital Filter

Abstract

The paper presents a theoretical analysis of the method that enables optimisation of the transient response of a digital filter or of an arbitrary discrete system, by pre-setting the initial conditions of the inner state description. For a particular design of a filter, it is enough to once evaluate coefficients, multiply them by the magnitude of the first sample of the signal and do filtration by using these initial conditions. The method was verified using the simulated, and NMR signals to maintain spectrum baselines correction, and will also be used for the study of an optimum filtration of the NMR signal with a variable instantaneous frequency.