Search

Seleno-Yeasts (SY) used as feed additives are known to contain different Selenium (Se) species. Seleno-Yeasts has been shown, on previous analytical methods, to contain selenomethionine (SeMet), selenocysteine (SeCys), selenate (SeIV) and selenite (SeVI), and various other organic and inorganic Se forms identified but rarely quantified. A new advanced method has allowed elemental Se (Se0), an inorganic Se species, to be quantified, thereby obtaining better insight into the proportion of inorganic Se in SY products. The study aimed to quantify the Se0 in SY products and assess the proportion of inorganic Se in SY. The Se speciation of 13 fresh commercials SY from different suppliers and batches, was assayed for the total Se, inorganic Se species (SeIV, SeVI and Se0), and organic Se species (SeMet and SeCys). Results on total Se were in line with the expected Se concentrations for all evaluated samples. The proportion of Se present as Se0 ranged from 3.6% to 51.8%. The quantity of Se0 in the SY products, added to SeIV and SeVI, indicated an average proportion of inorganic Se of 14.2% for the 13 analyzed SY products. The proportion of Se as SeMet ranged from 19.0% to 71.8%, (average of 55.8%), and a large variability in the SeMet content was observed. The SeCys content was also variable, with an average of 3.8%, relative to the total Se. In conclusion, advances in the analytical characterization have revealed that SY products can have a significantly high proportion of inorganic Se, which could affect the bioavailability of Se from SY supplements and explain their variable and lower bio-efficacy than pure SeMet supplements, such as hydroxy-selenomethionine.

In the shadow of the coronavirus (Covid-19) pandemic, the sterilization has become a major necessity for humans to avoid exponential Infection. In order to avoid touching contaminated surfaces, we develop an embedded system prototype, which allows the bottle or any other medium to pour the sterilization product without touching it. The human hands are detected using two sensors HC-SR04 and LM35. To overcome the stability problem due to the problem of sensor noise and enhance the system performance a Kalman filter algorithm is implemented to ensure stable hands detection. The efficiency of the prototype mounted on an Arduino board is checked. After the completion of the prototype, a comprehensive cost analysis is conducted.

The present article aims to investigate and evaluate the performances of DWDM (or Dense Wavelength Division Multiplexed) system under MIMO-FSO (or multiple input multiple output-free space optical) channel. However, the proposed system is configured and simulated in the worst-case meteorological conditions, and we base here on the variation of the attenuation coefficient. In this article, the proposed system is devoted to 90 users, every user exploits 30 Gbps. So, the total capacity of system is 90x30 Gbps. In addition, the use of MIMO technique in FSO channel ensures us the stability of the proposed system in the worst-case meteorological conditions and especially when the system exploits a high throughput. To assess the acceptable performance levels of the proposed system, it is necessary to use the following criteria: the minimum bit error rate (or BER) performance, the maximum quality factor (or Q-Factor) and the optical signal to noise ratio (or OSNR) performance. Finally, the proposed system in this paper provides a promising solution for 5G networks as back-up optical link.

Pending the arrival of the next generation of 5G which is not yet deployed insome countries like Algeria, 4G LTE remains one of the main mobile networks to ensure adequate quality services. Mostly, the deployment of femtocells to support the macrocell structure is crucial in the handover decision process. This paper presents a new approach called the epsilon Kalman Filter with normalized least-mean-square (ϵKFNLMS) to realize the handoff triggering in two-tier long-term evolution networks to ensure communication continuity to the pedestrian UE and improve mobility management. ϵKFNLMS uses a two-step process: a tracking process and a prediction process, to produce an optimal future state estimate at ”t+p”, where ”p” is the prediction footstep. The tracking process is performed by the Kalman filter, known for its precision in the state of the signal at time ”t”. It perfectly reduces the estimation error, injected afterward in the variable step-size NLMS algorithm (VSS-NLMS). While the predictionprocess is performed by the VSS-NLMS algorithm, an adaptive filterknown for its prediction of the future state at ”t+p”. Thus, the goal is to achieve a faster convergence with a steady-state. ϵ value provides a precise setting of the handover trigger. Through different numerical simulations in several indoor environments, the results show that the performance and effectiveness of the proposed approach (ϵKFNLMS) provide lower mean square error (MSE), stable physical appearance in the prediction process (convergence with a steady-state), and excellent speed of convergence compared to the classical Normalized LMS (NLMS) and Li-NLMS adaptive filters.

Simple Summary In recent years, interest has grown in improving the productivity of pure local breeds in Spain using corn silage as an alternative to concentrates. The aim of this investigation was to evaluate the lipid oxidation, color and visual color acceptability of meat from the Avileña-Negra Ibérica breed fed with corn silage in total mixed ration. The inclusion of corn silage led to lower lipid oxidation and higher visual color acceptability when the samples were packaged in a modified atmosphere. The use of corn silage could serve as a feeding option for this type of animal without affecting the quality of the derived meat. Abstract In order to increase the economic profitability of Avileña-Negra Ibérica beef production, the inclusion of corn silage in total mixed rations was proposed during the fattening period. Twenty-four Avileña-Negra Ibérica breed entire young bulls were used to evaluate the effects of two feeding systems—concentrate (CON) and corn silage (SIL)—and two packaging systems—vacuum (VAC) and modified atmosphere (MAP)—on the lipid oxidation, color evolution and visual color acceptability of meat throughout display with either oxygen-permeable film (FILM) or modified atmosphere (MAP). Two muscles were assessed: Longissimus thoracis (LT) and Semitendinosus (ST). Meat from SIL-fed animals had higher TBA values than CON-fed animals after 7 days in MAP packaging. Lipid oxidation was prevented more effectively by VAC packaging than MAP. Display time highly influenced lipid oxidation, since oxidation rates were lower with VAC than MAP packaging. After 14 days of display in MAP, the LT from CON was lighter than the ST. Meat discoloration after 7 days of display was significantly different between packaging systems. Vacuum-packaged meat kept the same color throughout the storage time. Visual color acceptability for the LT was higher throughout display than for the ST. Samples from the MAP were well accepted by consumers, especially the meat from the SIL group. Alternatively, feeding with corn silage could be used for this type of animals without affecting important aspects of meat quality.

International audience; Over the last few years, networks' infrastructures are experiencing a profound change initiated by Software Defined Networking (SDN) and Network Function Virtualization (NFV). In such networks, avoiding the risk of service degradation increasingly involves predicting the evolution of metrics impacting the Quality of Service (QoS), in order to implement appropriate preventive actions. Recurrent neural networks, in particular Long Short Term Memory (LSTM) networks, already demonstrated their efficiency in predicting time series, in particular in networking, thanks to their ability to memorize long sequences of data. In this paper, we propose an improvement that increases their accuracy by combining them with filters, especially the Fast Fourier Transform (FFT), in order to better extract the characteristics of the time series to be predicted. The proposed approach allows improving prediction performance significantly, while presenting an extremely low computational complexity at run-time compared to classical techniques such as Auto-Regressive Integrated Moving Average (ARIMA), which requires costly online operations.

Conférence virtuelle; International audience

In cognitive radio networks (CRNs), the topic of spectrum sensing (SS) or primary signal strength (PSS) has attracted a great deal of interest from academics and researchers to protect the legacy users spectrum. However, sensing enhances the robustness and reliability of spectrum management in CRNs environments. In order to improve the SS performance, the method used by secondary user (SU) or cognitive radio user (CRU) should be robust to analyze the transmitted signal strength by primary user (PU) or base station of primary networks (BSPNs). The spectrum sensing policy in the design of CRN systems is an important function, it determines the quality and performance of the system. In this context, we use the digital energy detector (DED) method which is employed by SU to sense the presence/or absence of the local PSS via two different channels, namely: The AWGN-link channel and the MRF-link (Multipath Rayleigh Fading) channel on two principal scenarios for the CRNs, a non-cooperative spectrum sensing (NCPSS) and a cooperative primary spectrum sensing (CPSS). Simulation results are investigated in terms of the detection probability (DP), false alarm probability (FAP) and the received signal-to-noise ratio (SNR) at SU.

The present article aims to investigate and analyze the performance of a 90 wavelengths multiplexed in optical link network using Dense Wavelength Division Multiplexed (DWDM) technology under a 20x20 multiple input multiple output (MIMO) based free space optical (FSO) communication system. The data rate for each wavelength is 30 Gbps, and the proposed scheme is simulated with the worst case, which is fog attenuation. The use of MIMO system ensures the data transmission with less bit error rate and improves the quality factor (Q-Factor) and optical signal to noise ratio (OSNR) of optical link. Finally, the suggested scheme provides a promising solution for the future as back-up optical links.

Recently, the issue of spectrum sensing (SS) in cognitive radio networks (CRNs) has been widely reported in the literature. This paper presents a new strategy to improve the reliability of SS. One of the challenges in CRNs is to sense all information about a secondary user (SU) that is moving, especially in a cluttered environment, because the mobility of that user has a considerable impact on the sensing performance of the primary user. Thereby, an algorithm that ensures a high sensing level, in the case of a SU moving with a low velocity in a cluttered environment, is proposed. Besides, the shadowing element becomes important in cluttered environments. Therefore, it is important to provide a weighted averaging mechanism to obtain a stable measure of the signal strength at the SU. To do this, a simple recursive estimator is used. In the end, the algorithm proposed is evaluated through simulations and results.

Mobile communication systems present an actuality subject in academic and industrial research activities due to several phenomena such as interferences, multipath, fading, and shadowing. All this lead to a severe perturbation on handover mechanism which depends on specific reports, essentially, reference signal received power (RSRP) and signal-to-interference and noise ratio(SINR). In this paper, we design a new technique in handover domain; it consists of combining energy detection method used in cognitive radio with least mean square (LMS) process in order to prognosticate the handover impact in a realistic scenario of heterogeneous LTE network. More exactly, technique sense of the word "triggering" will be changed to a probability of detection Pd. The proposed algorithm cycle follows two main steps; Firstly, predict at what time the absence of spectrum (primary user) will occur, using a predicted sensing probability Pˆd (t+p) . Secondly, search others spectrums in this time by calculating Pd(t) for each sensed signal and hand-off secondary user in the best spectrum. The results achieved of the simulation are evaluating, it shows that the proposed method predict the original Pd correctly with minimal errors and select the best spectrum successfully.

International audience; Low circulating levels of total and free 25-hydroxyvitamin D (25(OH)D) indicative of vitamin D status have been associated with obesity in humans. Moreover, obesity is thought to play a causal role in the reduction of 25(OH)D levels, and several theories have been put forward to explain this relationship. Here we tested the hypothesis that obesity disrupts vitamin D homeostasis in key organs of vitamin D metabolism. Male C57BL6 mice were fed for 7 or 11 weeks on either a control diet (control, 10% energy from fat) or a high-fat diet (HF, 60% energy from fat) formulated to provide equivalent vitamin D3 intake in both groups. After 7 weeks, there was a transient increase of total 25(OH)D together with a significant decrease of plasma vitamin D3 that could be related to the induction of hepatic genes involved in 25-hydroxylation. After 11 weeks, there was no change in total 25(OH)D but a significant decrease of free 25(OH)D and plasma vitamin D3 levels. We also quantified an increase of 25(OH)D in adipose tissue that was inversely correlated to the free 25(OH)D. Interestingly, this accumulation of 25(OH)D in adipose tissue was highly correlated to the induction of Cyp2r1, which could actively participate in vitamin D3 trapping and subsequent conversion to 25(OH)D in adipose tissue. Taken together, our data strongly suggest that the enzymes involved in vitamin D metabolism, notably in adipose tissue, are transcriptionally modified under high-fat diet, thus contributing to the obesity-related reduction of free 25(OH)D.

With the explosion use of multimedia in this millennium, image compression became indispensable for information technology users. It is a class of techniques that comes to efficiently store and transmit images while maintaining the best possible quality. Several software and methods respond to this problem of compression by maintaining a good balance between compression ratio and reconstructed image quality. In this paper, we present an approach that uses trigonometric functions combined with the famous DCT transform. Our aim is to utilize the arcsine reductive property and explore it better to gain more compression. Our experimental results showed a clear enhancement over the original DCT. Moreover, our method outperforms JPEG and JPEG2000 regarding compression ratio as well as the PSNR.

Image compression has wide area in image processing researches. It consists of minimizing the size in bytes of an image without degrading the quality to an unacceptable level. The reduction in file size allows more images to be stored in a given amount of disk or memory space. For this reason, many researches and compression techniques have been proposed; some of these have been effective in some areas and failed in others. In this paper, we propose a novel approach based on logarithm transform with pixels shifting. Using a non-uniform quantization technique, this approach can adapt to the type of image, making it more robust than existing methods. To assess the usefulness of our approach, we evaluated its performance on several benchmark images, and compared it to that of state-of-the-art methods based on Discrete Cosine Transform and Discrete Wavelet Transform. Results of these experiments showed our approach to be more accurate than existing methods on the tested images.

Handover is a process that allows a mobile node to change its attachment point. A mobile node connected to a network can, in order to improve the quality of service, have the need to leave it to connect to a cell either of the same network or of a new network. The present paper introduce three techniques using adaptive Variable Step-Size Least Mean Square (VSSLMS) filter combined with spectrum sensing probability method to detect the triggering of handover in heterogeneous LTE networks. These techniques are Normalized LMS (NLMS), Kwong-NLMS and Li-NLMS. The simulation environment is composed of two femtocells belonging to a macrocell. Five User Equipements (UEs) are positioned in one femtocell and are assumed closest to its circumference. Simulation results show that sensing probability with Li-NLMS algorithm has a better performance compared with classical NLMS and Kwong-NLMS.