RESUMEN
In the quantitative assessment of a system, a description of the low-order transfer function model is important. The objective of this study was to identify the system of a mechanomyogram (MMG) with SubSpace-based State Space model IDentification (4SID). The input data consisted of the electrical stimulation of the common peroneal nerve, which made the anterior tibial muscle contract. The output data consisted of the evoked MMG. We applied Fourier transform to the MMG signal and obtained a power spectrum. The 10th-order model was estimated by the 4SID method. It was suggested that the frequency band separation of the power spectrum reflected the types of recruited muscle fiber. The results suggest that the MMG is a linear system which can be estimated in the lower-order transfer function model by applying the 4SID to each frequency band.