Earthquake early-warning system (EEWS) is inevitable for saving human lives. The fast determination of the Earthquake’s (EQ’s) magnitude and its location is significant in disaster management and EQ risk mitigation. These parameters can be conveyed over the Internet-of-Things (IoT) network to alleviate an EQ disaster. In this article, a deep learning model based on integrating autoencoder (AE) and convolutional neural network (CNN) for a swift pinpointing of EQ magnitude and location after 3 s from the onset of the P-wave is proposed. Thus, we name it 3 s AE and CNN (3S-AE-CNN). The employed data set is observed by three stations from the Japanese Hi-net seismic network. We have trained our model on 12200 events (109.80 thousand 3-s-three-component seismic windows). The model facilitates the extraction of waveforms’ significant features leading to robust estimation of the EQ parameters. The proposed model predicts the magnitude and location of EQ with errors in magnitude, latitude, and longitude that reach 0.000028, 0.0000033, and 0.0001, respectively. The EQ’s parameters calculated by the proposed 3S-AE-CNN model are swiftly sent to a centralized IoT system that in turn directs the involved entity to take suitable action. The obtained results of the 3S-AE-CNN are compared to the conventional manual solution method, which represents the optimum solution mean. The 3S-AE-CNN shows an enhanced performance for the magnitude and location determination as compared with the benchmark method, which proves its effectiveness for EEWS.

Potential earthquake precursors include, among others, electromagnetic fields, gas emissions, Land Surface Temperature (LST), Sea Surface Temperature (SST), and Surface Air Temperature (SAT) anomalies. These observables have been individually studied, before earthquakes, by many researchers. The ionospheric studies concerning earthquakes (EQs) using magnetic data from Low Earth Orbit (LEO) satellites are increasingly being used to detect ionospheric anomalies before large EQs. Also, LST, SST, and SAT values retrieved from Moderate Resolution Imaging Spectroradiometer (MODIS) Terra and Aqua satellites and Modern-Era Retrospective analysis for Research and Applications Version 2 (MERRA-2) are considered as physical precursors before EQs. In this work, we jointly analyze magnetic, MODIS, and MERRA-2 data in space and time around the epicenters before the selected EQs in Mexico, Japan, Chile, and Indonesia. Our analyses present interesting findings where anomalies in temperature and magnetic field, preceding the considered EQs, are confirmed through different methods. Particularly, we utilize the Fast Fourier Transform (FFT) and the Discrete Cosine Transform (DCT) for analyzing magnetic data over the designated EQs regions. We use the magnetic data acquired by Swarm satellites in the top side ionosphere along with MODIS and MERRA-2. Five case studies are described to prove the effectiveness of our analyses. Precursory anomalies were observed using these methods in different anomalous days from the considered four regions of interest around the epicenter. It is concluded that these methods could be effective and reliable in detecting anomalies preceding the upcoming EQs.

In this paper, we employ a user-centric (UC) cell-free massive MIMO (CFmMIMO) network for providing ultra reliable low latency communication (URLLC) when traditional ground users (GUs) coexist with unmanned aerial vehicles (UAVs). We study power control in both the downlink and the uplink when partial zero-forcing (PZF) transmit/receive beamforming and maximum ratio transmission/combining are utilized. We consider optimization problems where the objective is to maximize either the users’ sum URLLC rate or the minimum user’s URLLC rate. The URLLC rate function is both complicated and nonconvex rendering the considered optimization problems nonconvex. Thus, we propose two approximations for the complicated URLLC rate function and employ successive convex optimization (SCO) to tackle the considered optimization problems. Specifically, we propose the SCO with iterative concave lower bound approximation (SCO-ICBA) and the SCO with iterative interference approximation (SCO-IIA). We provide extensive simulations to evaluate SCO-ICBA and SCO-IIA and compare UC CFmMIMO deployment with traditional colocated massive MIMO (COmMIMO) systems. The obtained results reveal that employing the SCO-IIA scheme to optimize the minimum user’s rate for CFmMIMO with MRT in the downlink, and PZF reception in the uplink can provide the best URLLC rate performances.

This article proposes deep learning (DL) framework constructed using deep autoencoder (DAE) to detect the malicious nodes in an Internet of Things (IoT) network assisted by the cognitive radio (CR) technology. In the IoT era, a plethora of nodes are connected to the network for the purpose of collecting and exchanging data. CR technology finds its role in IoT applications because of its ability to efficiently exploit the available spectrum. In this article, we consider IoT nodes as secondary users that perform cooperative spectrum sensing (CSS). Specifically, these IoT nodes sense the spectrum and send their reports to the fusion center (FC) to determine whether the spectrum is occupied by the primary user or not. In such a scenario, the existence of malicious IoT nodes might mislead the FC. To determine which of these IoT nodes are benevolent and which are malicious is crucial. We adopt a DAE-based DL framework, called DAE-TRUST, to detect malicious nodes in CR-assisted IoT. The proposed DAE-TRUST is able to identify the malicious nodes, whose reports can be excluded from the spectrum detection process. Simulation results, in six different real-world environments, show that our DAE-TRUST enhances the performance of CSS in IoT applications.

Ground vibrations caused by blasting operations in cement canisters is among the main mining issues that cause significant disruptions to nearby buildings and infrastructure. This research was performed in a limestone quarry situated southeast of Helwan City, Egypt, to investigate the impact of ground motion vibration due to cement blast action in limestone rocks. To reduce the environmental impact of quarry blasting, continuous monitoring, and accurate medium's peak particle velocity (PPV) assessment are required. Recently, machine learning (ML) models are employed in diverse applications. The default hyperparameters of such models must be modified to fit the problem concerned. The hyperparameters optimization for ML models impacts the employed model's performance and efficiency. In this research, different regression models are implemented for predicting the PPV values. A dataset representing 1438 blast incidents in the Helwan area was built and utilized to evaluate the considered ML models. This dataset incorporates the relationship of the ground vibration amplitude to both the explosive charge weight per delay and distance from the blast. The predictive models' output performance has been evaluated using the root-mean-squared error (RMSE) and the coefficient of determination ($R^{2}$ ). The PPV dataset has been divided into training and testing data to produce statistically significant results and to make the dataset more representative to avoid overfitting. The utilized test PPV dataset acts as a proxy for any new PPV data prediction. There was evidence of higher performance in the developed Decision Trees model with the lowest RMSE and the highest $R^{2}$ on training and testing data. The decision tree is, therefore, an acceptable algorithm for the construction of a predictive PPV model for other quarry blasting areas with conditions identical to those in Helwan. Finally, comparative experimental results have shown that optimized models can predict PPV values with lower errors and greater prediction accuracy.