Study for the Minimization of the Number of Frequencies for Cole-Cole Model for Bioimpedance Spectroscopy

Wearable bioimpedance devices are a growing trend toward low-power and portable devices. These devices offer many advantages, including portability, convenience, and integration into daily life. However, wearable BIS devices face power consumption issues. Reducing the frequencies in the stimulation signal has been proposed to address this issue. This study investigates the effect of minimizing the number of excitation signals on BIS performance using synthetic data from a Cole-Cole model. In the absence of noise, the simulation indicates that two frequencies are sufficient to reconstruct the impedance of a dispersion. When simulating noise variance levels of 10%, 5%, and 1%, the accuracy of component values and their relationship to frequency is affected. It should be noted that beyond a certain threshold, the results or component values are not significantly improved in the presence of noise by increasing the number of frequencies. However, when dealing with noise levels up to about 1%, it is critical to use at least three frequencies to accurately reconstruct the impedance spectrum. This approach reduces power consumption while maintaining the accuracy of the BIS measurements.