The fourth industrial revolution has transformed the industry, with information technology playing a crucial role in this shift. The increasing digitization of industrial systems demands efficient sensing and control methods, giving rise to soft sensors that have the potential to replace traditional physical sensors in order to reduce costs and enhance efficiency. This study explores the implementation of an artificial neural network (ANN) based soft sensor model in a water supply system to predict flow rates within the system. The soft sensor is centered on a Long Short-Term Memory (LSTM) artificial neural network model using Monte Carlo dropout to reduce uncertainty and improve estimation performance. Based on the results of this work, it is concluded that the proposed soft sensor (with Monte Carlo dropout) can predict flow rates more precisely, contributing to the reduction in water losses, as well as cost savings. This approach offers a valuable solution for minimizing water losses and ensuring the efficient use of this vital resource. Regarding the use of soft sensors based on LSTM neural networks with a careful choice of Monte Carlo dropout parameters, when compared to the multilayer perceptron model, the LSTM model with Monte Carlo dropout showed better mean absolute error, root mean square error, and coefficient of determination: 0.2450, 0.3121, and 0.996437 versus 0.2556, 0.3522, and 0.9954. Furthermore, this choice of Monte Carlo dropout parameters allowed us to achieve an LSTM network model capable of reducing uncertainty to 1.8290, keeping the error metrics also at low levels.

The distribution of water in urban areas involves several challenges, such as maintaining pipelines, controlling pressure and flow, and monitoring water quality. In particular, the measurement of the flow rate and pressure in pipelines is essential for optimizing water distribution in cities. In recent decades, new technologies have been used to address these challenges, such as hydraulic modeling systems with software, smart sensors, and automated control systems. Among the new possibilities, the use of wireless sensor networks has been highlighted. In this sense, IoT-based nodes have been proposed as a low-cost alternative, with the ability to communicate over the Internet with low energy consumption. Thus, this work describes the necessary steps, challenges, and solutions for the development of an autonomous IoT node applicable to monitoring pressure and flow in a water supply network. In the second part of the work, the data collected by the IoT nodes was processed to eliminate outliers and used to train a model based on artificial neural networks that are capable of predicting the flow in the system under monitoring. The results show that, based on the data measured by the proposed IoT node, it is possible to predict the flow in distribution systems operating in real time.

The cell-free (CF) systems are a key technology for future wireless networks, which exploit the scenario spatial macro-diversity to offer high channel capacity and link reliability, uniform user quality of service (QoS), and better coverage. A technique for access point selection (APS) must be applied to improve the use of the macro-diversity. This paper proposes a low-complexity APS technique based on the derived signal-to-interference-plus-noise ratio (SINR) theoretical expression. Additionally, to increase the system’s spectral efficiency (SE), an expression for the bit error probability (BEP) of the generic square M_m -ary quadrature amplitude modulation ( M_m -QAM) scheme for each user is proposed and applied to an adaptive modulation (AM) algorithm. The results revealed that the proposed APS technique outperforms techniques with similar simplicity and that the AM algorithm provides higher system SE maintaining the error rate below a given threshold.

This paper analyzes for the first time how the Beckmann distribution can be used to characterize the random variable that represents the envelope of the effective channel gain experienced by the k-th user equipment (UE) of a user-centric (UC) cell-free (CF) system in a scenario with noisy channel state information (CSI) estimation and line of sight (LoS). Additionally, it is shown how the Beckmann probability density function (PDF) can be used to derive the PDF and the cumulative density function (CDF) of the instantaneous signal-to-interference-plus-noise ratio (SINR) of the UC CF k-th UE, followed by applications in the ergodic capacity (EC) and outage probability (OP) expression derivations. It is shown that, regardless of the type of distribution considered for the channel gain between each access point (AP) and UE links, the effective gain presents a Beckmann distribution.