Fuzzy Algorithm for the EEG Signal Evaluation

The aim of the present paper is to propose an alternative technique based on the fuzzy methodology to investigate the dimensional complexity of biomedical signals generated by BCI. The signal identification is based on vectorial multifractal analyse that is mathematically formalized by the spectrum singularities of Hölder exponents evaluated on the interval [$0, \boldsymbol{T}$]. A fuzzy technique is used to calculate the Hölder exponents of the signal. The Husdorff parameters are evaluated by several techniques: averaging, norm, Center of Gravity Method and spectral estimation. A Type-2 Fuzzy Set method is used to classificate and sorte the signals. Numerical results of the propsed techniques are discussed.