Underwater wireless sensor networks (UWSNs) is an emerging technology for exploration of underwater resources. Security plays an important role in the UWSNs environment because the environment of UWSNs is prone to different security attacks. This research proposes SEECR: Secure Energy Efficient and Cooperative Routing protocol for UWSNs. SEECR comprised of energy efficient and strong defense mechanism for combatting attacks in underwater environment. SEECR exploits cooperative routing for enhancing the performance of network. Considering the resource constrained UWSNs environment minimum computation is employed for implementing security so that SEECR remains suitable for underwater environment. In order to evaluate the performance of SEECR, this research compares the performance of SEECR with AMCTD: Adaptive Mobility of Courier Nodes in Threshold-optimized DBR - a well-known routing protocol for UWSNs environment. The performance of SEECR and AMCTD protocols are evaluated using different performance evaluation parameters such as number of alive nodes, transmission loss, throughput, energy tax and end-to-end delay. The results suggest an improved performance of SEECR over AMCTD. SEECR shows an improvement of 9% in terms of number of alive nodes, over 50% reduction in terms of transmission loss, up to 9% increase in throughput, up to 23% reduction in energy tax, and 25% reduction in end-to-end delay. Further, we observe that attack significantly degrades the performance of AMCTD whereas due to the embedded defense mechanism in SEECR the impact of attack is negligible.

The rapid expansion of the Web of Things (WoT) and the Internet of Things (IoT) has raised security issues, with Denial of Service (DoS) attacks becoming increasingly prevalent. So, the aim of this study is to identify the security concerns in the four architectural layers of the Web of Things, particularly DoS attacks. For this study, existing literature are identified using search queries, and approximately 80 of relevant primary papers published in the recent decade are obtained after a thorough review which helps in addressing our research questions. After finding the relevant primary studies, we applied strict quality evaluation criteria to verify that all studies are evaluated. In addition, a taxonomy of deep learning (DL) techniques is presented on the basis of literature analysis conducted in this research, which is then used to characterize the various security concerns that occur in IoT and WoT systems. The study also examines which DL approaches are used to detect DoS/DDoS attacks in IoT and WoT. Our findings indicate that the optimal form of Intrusion Detection System (IDS) for dealing with DoS attacks is a hybrid IDS, which uses both the signature-based and the anomaly-based IDS. Moreover, DL techniques such as, CNNs and LSTMs, produced excellent results but are still in the development stage in terms of scalability and practical use. This review further highlights the present state of security mechanisms and sets the basis for future research, with an emphasis on refining DL-based techniques and improving the scalability and adaptability of security systems for WoT networks.

Intelligent traffic management systems have become one of the main applications of Intelligent Transportation Systems (ITS). There is a growing interest in Reinforcement Learning (RL) based control methods in ITS applications such as autonomous driving and traffic management solutions. Deep learning helps in approximating substantially complex nonlinear functions from complicated data sets and tackling complex control issues. In this paper, we propose an approach based on Multi-Agent Reinforcement Learning (MARL) and smart routing to improve the flow of autonomous vehicles on road networks. We evaluate Multi-Agent Advantage Actor-Critic (MA2C) and Independent Advantage Actor-Critical (IA2C), recently suggested Multi-Agent Reinforcement Learning techniques with smart routing for traffic signal optimization to determine its potential. We investigate the framework offered by non-Markov decision processes, enabling a more in-depth understanding of the algorithms. We conduct a critical analysis to observe the robustness and effectiveness of the method. The method’s efficacy and reliability are demonstrated by simulations using SUMO, a software modeling tool for traffic simulations. We used a road network that contains seven intersections. Our findings show that MA2C, when trained on pseudo-random vehicle flows, is a viable methodology that outperforms competing techniques.

The Web of Things is an improvement on the Internet of Things (IoT) that incorporates smart objects into both the web architecture (application) and the internet (network). WoT applications are inescapable in residential homes and communities. The intent behind WoT applications is to increase sustainable development for reducing resource consumption. The Web of Things (WoT) aims to create a decentralized Internet of Things. Edge computing addresses IoT computing demands by reducing the escalation in resource congestion situations. In edge computing data is placed closed to the end users which diverts computation load from the centralized data centers. Furthermore, the dispersed structure balances network traffic and minimizes traffic peaks in IoT networks. Therefore, resulting in reducing transmission delays between edge servers and end users which improves response times for real-time WoT applications. Low battery supply to nodes with enough power resources can increase the lifespan of the individual nodes by moving processing and communication overhead from the nodes. This paper integrates WoT and edge computing and compares their functionalities. In addition, it demonstrates how edge computing enhances WoT performance and concentrates on transmission, storage, and computation aspects. Furthermore, for performance evaluation it categorizes edge computing based on different architectures. Moreover, the challenges of Web of Things and edge computing have been discussed in terms of bandwidth, latency, energy, and cost. Finally, advantages of the Web of Things and edge computing have been discussed.

Seismic activity prediction has been a challenging research domain: in this regard, accurate prediction using historical data is an intricate task. Numerous machine learning and traditional approaches have been presented lately for seismic activity prediction; however, no generalizable model exists. In this work, we consider seismic activity predication as a binary classification problem, and propose a deep neural network architecture for the classification problem, using historical data from Chile, Hindukush, and Southern California. After obtaining the data for the three regions, a data cleaning process was used, which was followed by a feature engineering step, to create multiple new features based on various seismic laws. Afterwards, the proposed model was trained on the data, for improved prediction of the seismic activity. The performance of the proposed model was evaluated and compared with extant techniques, such as random forest, support vector machine, and logistic regression. The proposed model achieved accuracy scores of 98.28%, 95.13%, and 99.29% on the Chile, Hindukush, and Southern California datasets, respectively, which were higher than the current benchmark model and classifiers. In addition, we also conducted out-sample testing, where the evaluation metrics confirmed the generality of our proposed approach.

Chest disease refers to a variety of lung disorders, including lung cancer (LC), COVID-19, pneumonia (PNEU), tuberculosis (TB), and numerous other respiratory disorders. The symptoms (i.e., fever, cough, sore throat, etc.) of these chest diseases are similar, which might mislead radiologists and health experts when classifying chest diseases. Chest X-rays (CXR), cough sounds, and computed tomography (CT) scans are utilized by researchers and doctors to identify chest diseases such as LC, COVID-19, PNEU, and TB. The objective of the work is to identify nine different types of chest diseases, including COVID-19, edema (EDE), LC, PNEU, pneumothorax (PNEUTH), normal, atelectasis (ATE), and consolidation lung (COL). Therefore, we designed a novel deep learning (DL)-based chest disease detection network (DCDD_Net) that uses a CXR, CT scans, and cough sound images for the identification of nine different types of chest diseases. The scalogram method is used to convert the cough sounds into an image. Before training the proposed DCDD_Net model, the borderline (BL) SMOTE is applied to balance the CXR, CT scans, and cough sound images of nine chest diseases. The proposed DCDD_Net model is trained and evaluated on 20 publicly available benchmark chest disease datasets of CXR, CT scan, and cough sound images. The classification performance of the DCDD_Net is compared with four baseline models, i.e., InceptionResNet-V2, EfficientNet-B0, DenseNet-201, and Xception, as well as state-of-the-art (SOTA) classifiers. The DCDD_Net achieved an accuracy of 96.67%, a precision of 96.82%, a recall of 95.76%, an F1-score of 95.61%, and an area under the curve (AUC) of 99.43%. The results reveal that DCDD_Net outperformed the other four baseline models in terms of many performance evaluation metrics. Thus, the proposed DCDD_Net model can provide significant assistance to radiologists and medical experts. Additionally, the proposed model was also shown to be resilient by statistical evaluations of the datasets using McNemar and ANOVA tests.