In several applications of bioimpedance spectroscopy, the measured spectrum is parameterized by being fitted into the Cole equation. However, the extracted Cole parameters seem to be inconsistent from one measurement session to another, which leads to a high standard deviation of extracted parameters. This inconsistency is modeled with a source of random variations added to the voltage measurement carried out in the time domain. These random variations may originate from biological variations that are irrelevant to the evidence that we are investigating. Yet, they affect the voltage measured by using a bioimpedance device based on which magnitude and phase of impedance are calculated.By means of simulated data, we showed that Cole parameters are highly affected by this type of variation. We further showed that singular value decomposition (SVD) is an effective tool for parameterizing bioimpedance measurements, which results in more consistent parameters than Cole parameters. We propose to apply SVD as a preprocessing method to reconstruct denoised bioimpedance measurements. In order to evaluate the method, we calculated the relative difference between parameters extracted from noisy and clean simulated bioimpedance spectra. Both mean and standard deviation of this relative difference are shown to effectively decrease when Cole parameters are extracted from preprocessed data in comparison to being extracted from raw measurements.We evaluated the performance of the proposed method in distinguishing three arm positions, for a set of experiments including eight subjects. It is shown that Cole parameters of different positions are not distinguishable when extracted from raw measurements. However, one arm position can be distinguished based on SVD scores. Moreover, all three positions are shown to be distinguished by two parameters, R0/R∞ and Fc, when Cole parameters are extracted from preprocessed measurements. These results suggest that SVD could be considered as an effective technique for overcoming the variability of bio-impedance spectroscopy measurements.