Application of the Least Squares Method for Signal Processing and Determination of Parameters in Mechanical Myotonometry

ABSTRACT In this paper an approach based on the least squares method for signal processing and calculation of the muscle viscoelastic parameters in mechanical myotonometry is presented. Considering the specific waveform of the acceleration signal presenting muscle reaction in response to a short mechanical impact a mathematical model for signal processing is proposed. The equations for calculation of signal parameters’ values are derived and the appropriate numerical methods for their solving are determined. The approbation of the proposed algorithm for signal processing based on the mathematical model is performed in MATLAB. In order to compare the accuracy of the proposed approach with the direct one simulated signals with added noise and interferences are used. The normalized root mean squared errors in determination of signal parameters are calculated for different values of noise and interferences. The simulation results prove the higher accuracy achieved in determination of signal parameters with the proposed approach in comparison with the direct one. In conclusion, the proposed method for signal processing of mechanical myotonometry signals gives better results and could be applied for all noise levels and especially for the measurements with lower signal-to-noise ratios.