A novel type2-fuzzy adaptive filter is presented, which uses the concepts of type2-fuzzy logic, for electrocardiogram signals denoising. Type2-fuzzy adaptive filter is an information processor where both numerical and linguistic information are used as input-output pairs and fuzzy if-then rules, respectively. The proposed approach is based on an iterative procedure to achieve acceptable information extraction in the case where the statistical characteristics of the input-output signals are unknown. The proposed filter is presented as a dual-layered feedback system. Each layer has different function, the first layer being the type2-fuzzy autoregressive filter model. The second layer being responsible for training the membership function parameters. The second layer adjusts the type2-fuzzy adaptive filter parameters by using a teaching learning-based optimization algorithm (TLBO), which will allow the reaching of the required signal reconstruction by decreasing the criterion function. The proposed filter is validated and evaluated through some experimentations using the MIT-BIH ECGs databases where various artifacts were added to the ECGs signals; these included real and artificial noise. For comparison purposes, both model and non-modelbased methods recently published are used. Furthermore, the effect of the proposed filter on the malformation of diagnostic features of the ECG was studied and compared with several benchmark schemes. The results show that the proposed method performs better denoising for all noise power levels and for a different criteria viewpoint.

A linear hybrid multi-user detector in synchronous direct sequence code division multiple access systems (DS/CDMA) is performed by using linear successive interference cancellation detector (SIC) and linear decorrelator detector in order to exploit the advantages offered by the two detectors. Its main objectives are to reduce the long detection delay times and to improve the performance compared to that of the conventional SIC. Finally, theoretical and simulation results are in perfect agreement demonstrating the effectiveness of the elaborated scheme.

We propose a new approach for modeling a High Electron Mobility Transistor (HEMT) using that of Gaussian Process Regression one (GPR), to improve the current-voltage characteristics of HEMT transistor for using in electronic and biological domain or any other domain that needs it. The study and development of a new Atlas Silvaco device are taking into account the impact of several geometric and electric parameters; we focus on the electrical performances of the double gate field plate In0.2Ga0.8As/Al0.3Ga0.7As HEMT including double heterostructure; we compare the numerical simulation using 2D Atlas Silvaco simulator with the extracted experimental results. Then we validate our model by GPR approach. The GPR approach opens promising opportunities for devices modeling without knowing too much the device physics properties. The obtained results give better performances which lead to fabricate devices with better electrical properties for promoting further investigation.

This paper presents a hybrid strategy combining compact analytical models of short channel Gate-All-Around Junctionless (GAAJ) MOSFET and metaheuristic-based approach for parameters optimization. The proposed GAAJ MOSFET design includes highly extension regions doping. The aim is to investigate the impact of this design on the RF and analog performances systematically and to show the immunity behavior against the short channel effects (SCEs) degradation. In this context, an analytical model via the meticulous solution of 2D Poisson equation, incorporating source/drain (S/D) extensions effect, has been developed and verified by comparing it with TCAD simulation results. A comparative evaluation between the proposed GAAJ MOSFET structure and the classical device in terms of RF/Analog performances is also investigated. The proposed design provides RF/Analog performances improvement. Furthermore, based on the presented analytical models, Genetic Algorithms (GA) optimization approach is used to optimize the design of S/D parameters. The optimized structure exhibits better performances, i.e., cut-off frequency and drive current are improved. Besides, it shows superior immunity behavior against the RF/Analog degradation due to the unwanted SCEs. The insights offered by the proposed paradigm will help to enlighten designer in future challenges facing the GAAJ MOSFET technology for high RF/analog applications.

Unlike traditional visual cryptography based on a threshold mechanism in which the secret image can be revealed by stacking k,(2≤k≤n) shadows or more. In our suggested (kk,nn) progressive secret sharing scheme with unexpanded pixel, we could obtain until n−1n−1 revealed secret images with progressive enhanced contrast in an increasing way from the lowest quality, in the case of stacking two shares, to a highest quality in the case of stacking all the nn shares. Our scheme not only has the advantage of the unexpanded shares, but also the use of very simple Boolean XOR and OR operations to recover the secret image perfectly. Moreover, the proposed scheme does not need any codebooks to construct shares. The experimental results indicate the privilege of our method compared with the other related work.

Electrocardiogram (ECG) signals are often contaminated with artefacts and noises which can lead to incorrect diagnosis when they are visually inspected by cardiologists. In this paper, the well-known discrete Fourier series (DFS) is re-explored and an efficient DFS-based method is proposed to reduce contribution of both baseline wander (BW) and powerline interference (PLI) noises in ECG records. In the first step, the determination of the exact number of low frequency harmonics contributing in BW is achieved. Next, the baseline drift is estimated by the sum of all associated Fourier sinusoids components. Then, the baseline shift is discarded efficiently by a subtraction of its approximated version from the original biased ECG signal. Concerning the PLI, the subtraction of the contributing harmonics calculated in the same manner reduces efficiently such type of noise. In addition of visual quality results, the proposed algorithm shows superior performance in terms of higher signal-to-noise ratio and smaller mean square error when faced to the DCT-based algorithm.