Sofía I. Martín-González

Learn More
We present a new method for improving the classification score in the problem of binary hypothesis testing where the classes are modeled by a Gaussian mixture. We define a cost function which is based on the Chernoff distance and from it a transformation matrix is estimated that maximizes the separation between the classes. Once defined the cost function we(More)
We address the problem of estimating the path loss factor and its integration in RSS-based localization algorithms with wireless sensor networks. We propose an algorithm that relies on a stochastic characterization of the uncertainties in the propagation model. Due to that the path loss factor is unknown and the localization is only based on RSS(More)
In this paper the permutation entropy (PE) obtained from heart rate variability (HRV) is analyzed in a statistical model. In this model we also integrate other feature extraction techniques, the cepstrum coefficients derived from the same HRV and a set of band powers obtained from the electrocardiogram derived respiratory (EDR) signal. The aim of the model(More)
We introduce a sleep apnea characterization and classification approach based on a Heart Rate Variability (HRV) feature selection process, thus focusing on the characterization of the underlying process from a cardiac rate point of view. Therefore, we introduce linear and nonlinear variables, namely Cepstrum Coefficients (CC), Filterbanks (Fbank) and(More)
We present a new method for DOA estimation that relies on the assumption that the sources can be modelled as ARMA. A matrix formulation is introduced in the frequency domain so that the available information from the sources can be incorporated thus allowing the DOA estimation with a few sensors and achieve good estimates with acceptable computational(More)
We address the problem of improving the classification score in Bayesian binary hypothesis tests. We propose two methods to perform a dimensionality reduction by means of a linear matrix transformation from an original feature space to a new one. The criteria of the two methods is expressed in two different cost functions. The first function is based on the(More)
  • 1