Error estimation of digital filters with arbitrary structure and arithmetic by simulation

For the error estimation by simulation a matrix descrip tion of time invariant, causal, linear, discrete time sys tems is used, that yields a simulation algorithm which is independent of the individual filter structure to be simulated. Various quantization characteristics can easily be used by a modular simulation concept. Since the quantization characteristic of the simulating computer may influence the desired quantization characteristic, it is necessary to suppress the quantization of the simulating computer. Examples demonstrate the applicability of the described simulation method in the field of round-off noise analysis.