Abstract— This paper proposes an improved method for describing digital filter structures as part of digital signal processing (DSP) systems, characterized by high information content of its models. The proposed approach is used to perform synthesis of graphical analytical models of DSP systems as well as their analysis by means of multilevel decomposition. All the system elements are presented in the form of functional structures, and the input counts processing – by the discrete system – is considered as the process of converting arguments by these functional structures. Moreover, graphical analytical models of basic structures of finite impulse response (FIR) and infinite impulse response (IIR) filters are developed and integrated with object-oriented programming language to be used in various projects in the form of snap-in modules. Additionally, the method, proposed in this paper is used for synthesising an optimized structure of a digital filter for calculating the mean and current values of determined periodic signal at the optimal discrete system’s speed and invariance to discretization frequency. The obtained practical results show that the proposed method can be used with a great confidence in synthesizing optimized structures of digital filters as well as in the optimization of DSP systems’ structures with the purpose of identifying the mean and RMS voltage values. The latter is an indicator that the proposed method can be applicable for various electronic systems.