Baseline wander (BW) is a common low frequency artifact in electrocardiogram (ECG) signals. The prime cause from which BW arises is the patient's breathing and movement. To facilitate reliable visual interpretation of the ECG and to discern particular patterns in the ECG signal, BW needs to be removed. In this paper, a novel BW removal method is presented. The hypothesis is based on the observation that ECG signal variation covaries with its BW. As such, the P, Q, R, S, and T peaks will follow the baseline drift. On this basis, the following proposition is true: a reliable approximation of the baseline drift can be obtained from the shape derived from the interpolation of one form of the ECG signal peak (peak envelope). The simulation was performed by adding artificial BW to ECG signal recordings. The signal-to-noise ratio, mean squared error, and improvement factor criteria were used to numerically evaluate the performance of the proposed approach. The technique was compared to that of the Hilbert vibration decomposition method, an empirical-mode decomposition technique and mathematical morphology. The results of the simulation indicate that the proposed technique is most effective in situations where there is a considerable distortion in the baseline wandering.

In this investigation a novel type-2 fuzzy model for electrophysiological signals is presented. It is based on interval type-2 fuzzy systems. This method can deal with the curve fitting and computational time problems of type-2 fuzzy systems. This approach will significantly reduce the number of type-2 fuzzy rules and simultaneously preserve the fitting quality. The proposed model comprises a parallel interconnection of two type-2 sub-fuzzy models. The first is the primary model, which represents an ordinary model with a low resolution for the electrophysiological signal under consideration, the second is a fuzzy sub-model called the error model, which represents uncertainty in the primary model. Identification is achieved by innovative metaheuristic optimisation algorithms. The method's effectiveness is evaluated through testing on synthetic and real ECG signals. In addition, a detailed comparative study with several benchmark methods will be given. Intensive computer experimentations confirm that the proposed method can significantly improve convergence and resolution.

Reduction of the Total Harmonic Distortion (THD) in multilevel inverters requires resolution of complex nonlinear transcendental equations; in this paper we propose a combination of one of the best existing optimized hardware structures with the recent firefly algorithm, which was used to optimize the THD, through finding the best switching angles and guaranteeing the minimization of harmonics within a user defined bandwidth. The obtained THD through the simulation of the thirteen-level symmetric inverter has been reduced down to 5% (FFT of 60 harmonics). In order to validate the simulation results, a thirteen-level symmetric inverter prototype has been made, and practically experimented and tested with different loads. Consequently, the measured THD with resistive load was 4.7% on a bandwidth of 3 kHz. The main advantage of the achieved work is the reduction of the THD.

A robust nonlinear controller based on an improved feedback linearization technique with state observer is developed for a class of nonlinear systems with uncertainties and external disturbances. First, by combining classical feedback linearization approach with a discontinuous control and a fuzzy logic system, we design and study a robust controller for uncertain nonlinear systems. Second, we propose an optimized extended Kalman filter (EKF) for the observation of the states. The parameters to be optimized are the covariance matrices Q and R, which play an important role in the EKF performances. The particle swarm optimization algorithm insures this optimization. Lyapunov synthesis approach is used to prove the stability of the whole control loop. The proposed approach is applied on a two-link robot system under Matlab environment. Simulation results have confirmed the effectiveness of the proposed approach against uncertainties and external disturbances; and exhibited a more superior performance than the non-improved control actions.

The aim of this paper is to examine the effectiveness of combining three popular tools used in pattern recognition, which are the Active Appearance Model (AAM), the two-dimensional discrete cosine transform (2D-DCT), and Kernel Fisher Analysis (KFA), for face recognition across age variations. For this purpose, we first used AAM to generate an AAM-based face representation; then, we applied 2D-DCT to get the descriptor of the image; and finally, we used a multiclass KFA for dimension reduction. Classification was made through a K-nearest neighbor classifier, based on Euclidean distance. Our experimental results on face images, which were obtained from the publicly available FG-NET face database, showed that the proposed descriptor worked satisfactorily for both face identification and verification across age progression.