Ensuring the secure communication of unmanned aerial vehicle-assisted reconfigurable intelligent surface (UAV-RIS) is crucial in maintaining seamless connection in next-generation Internet of Things (IoT) networks. For this purpose, intelligent beamforming is essential to ensure secure data transmission from IoT devices to UAV-RIS and optimize communication while preventing adversarial attacks. This paper proposes a novel framework of federated learning for long short-term memory-based deep deterministic policy gradient with an attention mechanism (F-DDPG-AM). The proposed algorithm aims to improve security and mitigate potential threats in UAV-RIS-assisted IoT networks. The F-DDPG-AM combines the federated LSTM’s power to capture long-term dependencies in sequential data with the attention mechanism to focus on key network states and improve decision-making efficiency. The F-DDPG-AM framework improves learning efficiency, accelerates convergence, and enhances resilience against adversarial attacks by selectively prioritizing crucial network information and focusing on insecure scenarios. In addition, federated learning in the proposal ensures secure decision-making through local training for UAV-RIS-enabled IoT networks. The F-DDPG-AM enhances system scalability, trustworthiness, and compliance with secure machine learning principles by decentralizing the training process. The simulation results demonstrate the superior performance of the proposed F-DDPG-AM framework in defending against attacks, significantly outperforming traditional security approaches and other existing reinforcement learning models.

Nowadays, object saliency detection has attracted considerable attention due to the vivid description of images. The traditional saliency detection methods are faced with the challenges of sparse boundary, fractured contour and internal non-uniform density. That will result in losing texture and detailed information of images, which makes a bad contribution to the subsequent object detection. Therefore, we propose a new approach based on fuzzy region competition and a Gaussian mixture segment model via a modified line segment detector (FGM-MLSD) for airport saliency detection in remote sensing images. First, we adopt fuzzy region competition, combining a Gaussian mixture model to segment the input images and obtain the airport candidate regions. After segmentation, a modified line segment detector (LSD) is used for extracting line features, which enhances the connection between broken lines and greatly improves the detected line quality. Then we can acquire the saliency map of the airport region. At last, we fuse the above saliency maps with the binarization map obtained by the Otsu method, aiming to eliminate the false alarm. Finally, abundant experiments are conducted, and the testing results reveal that the neoteric method can clearly and accurately extract the airport region in the remote sensing images and effectively improve the accuracy of saliency detection.

Underwater communication systems face challenges, including limited bandwidth, high latency, and void areas. This article introduces synergy optimization routing protocol (SORP) for Internet of Underwater (IoU) sensor networks, emphasizing link scheduling to address localization, energy consumption, latency, network longevity, and void regions. Leveraging belief-desire-intention (BDI) and fuzzy logic, SORP offers adaptive responses to varying network conditions. Theoretical modeling using NetLogo enhances the understanding of SORP’s behavior. In evaluation, SORP consistently outperforms others. Demonstrating superior energy efficiency (3.0–3.9) compared to PPWURC, state prediction-based data collection (SPDC), balanced routing protocol based on machine learning (BRP-ML) (40–120), and energy efficient clustering routing protocol based on arithmetic progression (5–6.5), SORP proves its efficacy. Latency analysis reveals SORP consistently displaying the lowest values (2.0–2.8), surpassing packet hierarchy and void processing, SPDC, and BRP-ML (8.3–30.0). With a perfect packet delivery ratio (PDR) of 98%, SORP showcases exceptional reliability. Network lifetime analysis positions SORP as a durable option, lasting from 3985 to 4010 rounds. Validation through an underwater communication system demonstrates speeds of 5 Mb/s and above. Simulation testing reveals a transmission speed of 80 bps with latency of less than 4 s and 98% PDR. Theoretical predictions indicate significant improvements in real-time transmission, reducing latency to less than 1 s with a speed of 5 Mb/s. This research presents an innovative and practical approach to address underwater communication challenges, highlighting the efficiency and reliability of SORP in routing protocols for underwater sensor networks. The combination of theoretical modeling and real-time testing offers a comprehensive understanding, emphasizing the potential real-world impact of SORP.

Multihop communication has gained prominence within the realm of the Internet of Underwater Things (IoUT) owing to its exceptional reliability amidst the challenges posed by the underwater acoustic environment. Despite this, the persistence of limitations caused by propagation delay, high collision rate, and limited energy in underwater communication remains, representing the most formidable hurdles in ensuring the successful transmission of data gathered by sensor nodes. To address these challenges, we employ a machine learning (ML)-based optimal cooperating node selection for each hop, considering the Shortest propagation delay, minimal residual Energy, and a low Collision rate (referred to as SEC). For this purpose, we initially assemble the sensor nodes to create a list of cooperative nodes, considering the aspect of SEC. Then, using an assembled list of cooperating sensor nodes, we employ ML-based algorithms, such as reinforcement learning (RL-SEC), deep Q-networks (DQN-SEC), and deep deterministic policy gradient (DDPG-SEC), to predict the optimal cooperating node for each hop. The simulation results of the DDPG-SEC demonstrate a significant improvement of approximately 56% when compared with RL-SEC, DQN-SEC, and other state-of-the-art techniques.

In recent years, driven by the continuous development of mobile Internet technology and artificial intelligence technology, the improvement of the manufacturing level of 6G Internet-of-Everything (IoE) products and the increase in residents’ income level, the 6G IoE industry has shown a sustained and stable development trend. However, 6G IoE has great security risks. Network anomaly detection is very important for 6G IoE. The anomaly detection method based on traditional deep auto-encoder uses the reconstruction error to determine whether the sample to be measured is normal data or abnormal data. However, the reconstruction errors generated by the above method on normal data and abnormal data are very close, which leads to some abnormal data being easily misclassified as normal data. Therefore, an anomaly detection method based on deep auto-encoder and capsule graph convolution via sparrow search algorithm in 6G IoE is proposed. Firstly, the capsule graph network uses the bottleneck feature of the input sample to generate the bottleneck feature of the pseudo-abnormal data, so as to increase the abnormal data information in the training set. The capsule dynamic fusion strategy aggregates different factors to obtain new item embedding. Secondly, deep auto-encoder reconstructs the bottleneck characteristics with abnormal data information into normal data as much as possible, and increases the difference of reconstruction error between abnormal data and normal data. In the process of network classification, we use the sparrow search algorithm to find the optimal value of the function. And at the same time, it prevents the algorithm from prematurity and improves the classification effect. Finally, we conduct experiments on public data sets to compare with other advanced methods. Experimental results show that the proposed method can effectively enlarge the difference between normal data and abnormal data in reconstruction error.

Recent advances in artificial intelligence have led to deepfake images, enabling users to replace a real face with a genuine one. deepfake images have recently been used to malign public figures, politicians, and even average citizens. deepfake but realistic images have been used to stir political dissatisfaction, blackmail, propagate false news, and even carry out bogus terrorist attacks. Thus, identifying real images from fakes has got more challenging. To avoid these issues, this study employs transfer learning and data augmentation technique to classify deepfake images. For experimentation, 190,335 RGB-resolution deepfake and real images and image augmentation methods are used to prepare the dataset. The experiments use the deep learning models: convolutional neural network (CNN), Inception V3, visual geometry group (VGG19), and VGG16 with a transfer learning approach. Essential evaluation metrics (accuracy, precision, recall, F1-score, confusion matrix, and AUC-ROC curve score) are used to test the efficacy of the proposed approach. Results revealed that the proposed approach achieves an accuracy, recall, F1-score and AUC-ROC score of 90% and 91% precision, with our fine-tuned VGG16 model outperforming other DL models in recognizing real and deepfakes.

This paper proposes a generation driven control strategy for handling the operations of a smart micro grid. This control strategy is proposed for a PV and battery based microgrid and tested using technical virtual power plant (TVPP) realization. Mesh network infrastructure is used to connect the generation and load elements. The projected strategy is simulated in MATLAB and tested on IEEE RTS96 73 bus system using Simulink. The advantages of VPP approach in operating and maintaining micro grids are discussed, and concluded that a TVPP realization helped in optimizing the operation of the grid in the all four quarters of the year. Hence; five load nodes were completely shut down during peak periods in Jan-Mar quarter and 6 nodes were operating at 75% of the required demand in that quarter.

This article proposes a face detection algorithm based on deep learning for mobile applications. Face detection is a pre-processing step for many high-end computer vision tasks. Therefore, outcomes of this task directly influence the accuracy of the other tasks, such as recognition, tracking, relighting, and other tasks. Face detection is a well studied research extent; however, it still endures from challenges such as face orientation, occlusion, lighting conditions, and alteration in facial landmarks. The projected technique aims at targeting such crucial challenges with light weight deep learning architecture. The proposed algorithm is tested on the publicly available perplexing datasets and has shown promising results. The effectiveness of the proposed algorithm is proved quantitatively and qualitatively with state of art techniques.