This article examines the influence of radio and traffic features on perceived downlink throughput in an Orange Senegal 4G LTE small-cell network, using a set of real data collected from seven small cells over a period of more than three months. The main objective is to identify the feature that best explains variations in perceived downlink throughput and to assess the performance of machine learning models (LR, DT, RF, and MLP) and a deep learning model (DNN) for its prediction. After rigorous data preprocessing and the selection of 13 relevant features, correlation and prediction analyses are conducted. The results show that, among all the features studied, the average Channel Quality Indicator (CQI_Avg) is the most decisive factor affecting perceived downlink throughput. This dominance is explained by the CQI’s direct causal role in link adaptation mechanisms (modulation and coding), in contrast to load or traffic features, and whose impact remains indirect. From a predictive standpoint, the deep neural network (DNN) outperforms all other models, achieving an accuracy of 96.1

Accurately predicting wireless channel quality is crucial for enabling mobile network operators to carry out proactive network operations. In this article, we address the challenge of predicting channel quality across various wireless links. Building on our previous work, we introduce two new models: a Convolutional Neural Network (CNN) and a hybrid CNN-LSTM, which combines CNN with a Long Short-Term Memory (LSTM) architecture. These models are designed to predict the Channel Quality Indicator (CQI) in 4G LTE/5G networks incorporating small cell base station architectures. We evaluate their performance using a dataset collected from Orange Senegal’s commercial 4G LTE/5G network. Our results demonstrate that the CNN, LSTM, and CNN-LSTM models adapt effectively to real-world conditions and achieve high prediction accuracy. Among them, the CNN-LSTM model delivers the best performance (RMSE = 0.25), followed by the LSTM model (RMSE = 0.281), and the CNN model (RMSE = 0.308).

Sahel is an African area with high solar potential.However, this potential is not uniform across the region.This paper examines the spatial distribution of the available solar potential by using six stations across the Sahel area.This comparative study was based on the analysis of in situ measurements in Dakar in Senegal, Niamey in Niger, Ouagadougou, Gaoua, Dori in Burkina Faso and N'Djamena in Chad.The results showed the presence of a good global solar potential with an average value of about 5.43 kWh/m 2 /day.The maxima of global potential are noted in the northern part in Niamey with a value of 6.24 kWh/m 2 /day while the minima are recorded in the south-eastern part in N'Djamena with an irradiation close to 4.71 kWh/m 2 /day.Then, the monthly evolution of this potential shows similar trends for all stations.Indeed, two maximums are observed during the year in Spring (March) and Autumn (October).However, for most of these stations, the minima of global potential are recorded in Winter (November, February) and during the rainy season (July, October).Moreover, the direct normal potential also shows seasonal trends for the two stations (Dakar, Niamey) where it was measured.The maxima of direct normal irradiation (DNI) are observed between February and May with a value of 5.5 kWh/m 2 /day in Dakar and in Niamey with a value around 5.32 kWh/m 2 /day between February and November.

Atmospheric distillation is the first step in separating crude oil into by-products. It uses the different boiling temperatures of the components of crude oil to separate them. But crude oil contains a large quantity of acids and corrosive gases, including sulfur compounds, naphthenic acids, carbon dioxide, oxygen, etc. However, the temperature has an important influence on the aggressiveness of the corrosion factors in the atmospheric distillation column. This paper aims to investigate the role of temperature on corrosive products in the atmospheric distillation column. The results of the developed model show that the temperature increases the corrosion rate in the atmospheric distillation column but above a certain temperature value (about 600 K), it decreases. This illustrates the dual role played by temperature in the study of corrosion within the atmospheric distillation column.

The massive growth of wireless traffic goes hand in hand with the deployment of advanced radio interfaces as well as network densification. This growth has a direct impact on the radio access architecture, which today is moving from centralized to distributed deployments through the use of a large number of access points (APs). This paper verifies the feasibility of deploying multiple APs in series on a single line in a ring topology in a cell-less network. On the one hand, this technique will further improve the communication capacity and flexibility of a Radio-over-Fiber (RoF) based mobile communication system and will reduce its construction cost. And on the other hand, this deployment topology is a solution to achieve a massive cell-free Multiple-Input Multiple-Output (MIMO) architecture and a cost-effective fronthaul solution. First, a passive optical add/drop multiplexer (OADM) is used to extract and add downlink and uplink signals from the remote access points of one kilometer. Then, a deployment model is developed with version 17 Optisystem software. The results obtained showed that the quadrature amplitude modulation (QAM) does not adapt to this multi-carrier transmission to deploy several AP in series on a single line. Thus, the performance degradation increases when the number of APs integrated on the line increases.

The study presented in this article focuses on the temporal dynamics of wind energy production at the Taïba Ndiaye wind farm in Senegal, with a capacity of 158.7 MW. The monthly and seasonal distribution of production shows a strong trend, with maximums recorded between December and May (winter and spring) at around 1800 MWh, and minimums between July and November (summer and autumn) with production below 500 MWh. The diurnal cycle representation exhibits variation with a marked cycle, particularly between November and April. Night-time production is higher than daytime production by more than 43%. The effects of 100-m wind on the farm production are also analysed and show a positive correlation between wind speed and production throughout the year. Production peaks observed in winter and spring are caused by strong winds (approximately 8.5 m/s), while the lowest levels recorded during the summer season are due to weather conditions characterized by weak winds (less than 4 m/s). Similarly, optimal wind directions are observed in winter and spring, periods of maximum production, when the winds blow between the northwest and northeast.

Prior knowledge of wireless channel quality with high accuracy is essential to enable anticipated networking tasks. Traditional channel quality prediction problems rely on past channel information to predict its future quality. In this paper, we investigate the channel quality prediction problem over different wireless channels. We propose an efficient prediction scheme based on deep learning, to predict channel quality. For the deep learning task, we use deep neural networks and long short-term memory networks. We compare their performance on a dataset collected from a commercial 4G mobile radio network of Orange Senegal. The performance evaluation performed on the benchmark dataset demonstrates the validity of the proposed deep learning approach, reaching a root mean square error of 0.27 for the LSTM model and 0.28 for the DNN model. The performances in terms of RMSE with the same dataset for each of the models used in this study were compared to other models. Thus, the DNN and LSTM models give low RMSEs compared to the models of our previous work. The proposed prediction method can be applied for 5G small cell networks.

Using a mobile dataset from Orange Senegal small cell 4G network, we study the effect of variables such as data traffic at the downlink level, data traffic at the uplink level, total data traffic, maximum number of active users, signaling protocol, uplink user rate, downlink user rate, physical resource block rate for downlink, block rate physical resources for uplink, load data logging, channel quality indicator, downlink radio delay average, over the perceived rate at the downlink. We are looking to find the variable that most affects the perceived flow. We do this by using machine learning to find the variable that closely explains the variation in perceived flow and helps predict flow with greater accuracy. We observe that correlation analysis is unable to find a hidden relationship between throughput and other variables. With models such as linear regression, decision tree, random forest and multi-layered perceptron, the channel quality indicator (CQI_Avg) turns out to be the variable that more closely explains the variation in perceived flow and more accurately contributes to the prediction compared to others variables.

In recent years, photovoltaic (PV) modules are widely used in many applications around the world. However, this renewable energy is plagued by dust, airborne particles, humidity, and high ambient temperatures. This paper studies the effect of dust soiling on silicon-based photovoltaic panel performance in a mini-solar power plant located in Dakar (Senegal, 14°42'N latitude, 17°28'W longitude). Results of the current-voltage (I - V) characteristics of photovoltaic panels tested under real conditions. We modeled a silicon-based PV cell in a dusty environment as a stack of thin layers of dust, glass and silicon. The silicon layer is modeled as a P-N junction. The study performed under standard laboratory conditions with input data of irradiation at 1000 W/m2, cell temperature at 25°C and solar spectrum with Air Mass (AM) at 1.5 for the monocrystalline silicon PV cell (m-Si). The analysis with an ellipsometer of dust samples collected on photovoltaic panels allowed to obtain the refraction indices (real and imaginary) of these particles which will complete the input parameters of the model. Results show that for a photon flux arriving on dust layer of 70 μm (corresponding to dust deposit of 3.3 g/m2) deposited on silicon-based PV cells, short circuit current decreases from 54 mA (for a clean cell) to 26 mA. Also, conversion efficiency decreases by 50% compared to clean cell and the cell fill factor decreases by 76% - 50% compared to reference PV cell.

Two dust samples were collected from the same solar panels on June, 2017 in Dakar. The first was sent to Ithemba Labs (South Africa) and the second to Lincon Nebraska University (USA) for structural and physical properties study. This work aims to evaluate the dust impact on PV efficiency. The study shows that these particles have a non-spherical shape with sizes between 0 and 50 μm. However, most of them have a size about 2 μm. Elemental composition based on Energy Dispersive X-ray microanalysis indicates that dust is dominated by elements such as O, Na, Mg, Al, Si, Cl, K, Ca, Ti, Mn and Fe. Some elements such as P, S, Zn, Sr, Zr and Cr are minority or in traces form. The analysis shows that this African dust is a mixture of different chemical compounds with a predominant phase consisting of SiO2 type quartz about 73.8% of the total and the second phase is certainly calcite(CaCO3) representing 13.6% of the collected particles. The rest (12.6%) is presumably a mixture of Iron Oxide Chloride (FeOCl), Mantienneite (Al2FeH33K0.5Mg3O34P4Ti) and Kaersutite (Al2Ca2Mg6NaO24Si6). Finally, the diffuse reflectance spectroscopy (DRS) shows that these particles reflect more than 70% of the irradiation reaching the PV panels surface.

This study aims to evaluate the optical losses of photovoltaic modules due to Saharan dust deposition in Dakar, Senegal, West Africa. For this purpose, an air-dust-glass system is modeled to simulate optical losses in transmittance and reflectance. To do this, we have collected dust samples from Photo-Voltaic (PV) surface in Dakar area (14°42'N latitude, 17°28'W longitude), Senegal. X-ray fluorescence reveals that silicon (Si), iron (Fe), calcium (Ca) and potassium (K) mainly composed these dust samples. Then, dust refractive indices obtained from an ellipsometer were used as an input to be used in the model. Simulations show that for radiation (at normal incidence) arriving on a dust layer of 30 μm-thick (corresponding to a dust deposit of 1.63 g/m2), 79% of the visible spectrum is transmitted; 19% is reflected and 2% is absorbed. Overall, the transmittance decreases by more than 50% as of dust layer of 70 μm-thick corresponding to a dust deposit of 3.3 g/m2.

This study aims to evaluate dust impact on climate parameters over the Sahel region by RegCM3 regional model during 2006. Indeed, aerosols are one of the main uncertainties in climate models. The aerosol optical depth (AOD) derived from RegCM3 model has been validated with various observed datasets. The aerosol sources are identified over North Algeria and East of Sahel (Bodele depression). Discrepancies are noted when considering dust temporal and spatial distribution. Dust season extends between March and October, with two peaks of AOD recorded in March (spring) and June (summer). The dust vertical distribution showed that the mineral aerosol layer is located between 850 hPa and 300 hPa (1.5 km to 7 km). The RegCM3 model simulates fairly well the transport in the upper layers, especially in the Saharan Air Layer (SAL) during the summer. However, RegCM3 simulates poorly the transport and sedimentation of particles in the lower layers (below 2 km). The investigation of dust radiative impact shows a general cooling. The maximum of radiative forcing is located around 18°N - 20°N, with values of about -80 W/m2 in June - August (JJA) and -40 W/m2 at the surface during March - May (MAM). This study also showed the indirect effect of dust with a decrease in precipitation about -0.7 mm/day around 15 - 20°N during the rainy season.

This study aims to evaluate the long-term variations of sunshine duration and to estimate its interaction with meteorogical parameters from 1950 to 2010 in Chad, Central Africa. The results show that Chad is the one of world’s sunniest countries. Each year, Chad receives more than 3030.91 ± 176.33 hours of sunning corresponding to 8.9 hours daily. Likewise, a strong north-south gradient is noted over Chad. For instance, the daily insolation is 10 ± 0.41 hours in the north, 8.85 ± 1.1 hours in the center and 7.75 ± 1.8 hours in the south. Furthermore, there is a marked seasonality of sunshine duration with maximums in dry season and minimums during the rainy season. The lowest values of sunshine duration are found in August. On the contrary, maximums are recorded from November to February with values greater than 9.5 hours per day. Moreover, the annual anomalies study allowed determining three great periods in terms of sunshine variability in Chad. Firstly, the period between 1950 to 1970 (named humid period) is characterized by the lowest values of sunshine duration in Chad. Secondly, from 1970 and 1990, the region suffered an unprecedented drought which resulted in an increase of sunstroke duration in Chad. And finally, the period from 1990 to 2010 called the return period is characterized by a rapid year-to-year fluctuations of insolation duration. Unlike to surface temperature, we have also shown that there is a direct relation between the duration of insolation and the meteorological parameters such as precipitations and relative humidity.

The objective of this work is to evaluate the available solar potential at N'Djamena (12°08N, 15°04E) from 2017 to 2018. To achieve this goal, we used various datasets and model including: the in situ shortwave radiation (by pyranometer) measurement and sunshine duration (by Campbell-Stokes heliograph) obtained from N'Djamena station, observations from MODIS (aerosol optical depth (AOD) and precipitable water) satellite sensors, and simulations from Streamer radiative code. The results show the presence of a good available solar potential with an annual global potential of 4.71 kWh/m2/d. At the intra-seasonal time scale, there are two maximums for the global solar potential. The first maximum is registered in the month of March (spring) with value of 5.7 kWh/m2/d and the second in October (autumn) with value of 5.18 kWh/m2/d. However, the minimum of global potential is recorded in winter (from December to February) with values around 3.86 kWh/m2/d. Then, the measured global irradiation allowed validating the Streamer radiative transfer code with a score of more than 98%. Subsequently, this model was used to simulate direct normal and diffuse irradiation for several types of days (clear, dusty and cloudy days). An examination of the dust influence on solar radiation based on selected cases (AOD = 2.05) indicates a mean decrease of 3.33 and 3.17 kWh/m2/d, respectively, for the total and direct normal potential. This corresponds to an increase of the diffuse potential of 0.52 kWh/m2/d. Finally, an increase of 5.82 cm of precipitable water per day tends to decrease the overall potential of 0.73 kWh/m2/d and the direct normal potential of 1.74 kWh/m2/d. For this cloudy day, the potential has increased more than 0.89 kWh/m2/d.

The objective of this work is to study the seasonal distribution of dust dry deposition in West Africa particularly in Senegal. This initiative is part of the efforts to improve the performance of photovoltaic panels (PV) in dusty environments such as Sahel region. It will help to evaluate the impact of dust dry deposition on these PV. Two climate models including dust modules during 2006–2010, were used: MERRA-2 (Modern-Era Retrospective Analysis for Research and Applications, version 2) reanalysis and ALADIN (Aire Limitée Adaptation dynamique Développement InterNational) model. In Mbour (Senegal), the aerosol optical depth (AOD) from these two products has been validated by in situ data. Indeed, MERRA-2 and ALADIN fairly simulate the AOD with a maximum in March and June. However, these products tend to overestimate measurements, especially for ALADIN. The correlation coefficient between in situ AOD and products is evaluate almost 0.83 for MERRA-2 and 0.72 for ALADIN. From 2007 to 2009, dust deposition measurement campaign was conducted in Mbour. The comparison between these seasonal data and simulations show a right coherence even if MERRA-2 underestimates measurement and ALADIN overestimates it. The correlation compared to in situ measurements is estimate to 0.71 for MERRA-2 and 0.72 for ALADIN. However, the results showed that ALADIN better describes the seasonal dry deposition during dry season which lasts 9 months despite a strong overestimation in winter. Finally, long-term simulation with ALADIN show that dry deposition maximum occurs from December to May in Senegal and throughout the West Africa region.