A Crest-Factor Optimization Algorithm for Multisine Signals based on the Evolutionary Role Playing Game Theory

In impedance spectroscopy, excitation signals should be limited in their amplitude to fulfil the linearity condition and in their duration to fulfill the stability condition. Multisine excitation signals are suitable because they can reduce the excitation time duration, but they need to be optimized concerning crest factor (CF). A higher challenge is the optimization of the crest factor of multisine signals. In this paper, we propose a hybrid deterministic-heuristic optimization method for CF optimization based on the Evolutionary Role Playing Game Theory able to realize multisine signals with lower and more consistent CF for logarithmically distributed signal frequency components. The novel optimization method needs a short execution time, which is comparable to that of deterministic methods. The method validation has been carried out for bioimpedance measurement. In the case of bioimpedance measurement, the new method reaches a CF with a median of 1.76, which is significantly lower than state-of-the-art methods realizing a CF typically between 1.89 and 2.79.