This paper proposes an innovative dual-polarized reconfigurable reflectarray element, which can continuously tune the phase shift for the dual-polarized reflecting waves. To achieve phase tuning capability, a varactor...
详细信息
Wireless node localization is a critical need for many underwater applications, that heavily depends on mobile node position estimation. Data is sent between nodes using three main methods of communication-i) acoustic...
详细信息
The ever-expanding Internet of Things (IoT) landscape presents a double-edged sword. While it fosters interconnectedness, the vast amount of data generated by IoT devices creates a larger attack surface for cybercrimi...
详细信息
The ever-expanding Internet of Things (IoT) landscape presents a double-edged sword. While it fosters interconnectedness, the vast amount of data generated by IoT devices creates a larger attack surface for cybercriminals. Intrusions in these environments can have severe consequences. To combat this growing threat, robust intrusion detection systems (IDS) are crucial. The data comprised by this attack is multivariate, highly complex, non-stationary, and nonlinear. To extract the complex patterns from this complex data, we require the most robust, optimized tools. Machine learning (ML) and deep learning (DL) have emerged as powerful tools for IDSs, offering high accuracy in detecting and preventing security breaches. This research delves into anomaly detection, a technique that identifies deviations from normal system behavior, potentially indicating attacks. Given the complexity of anomaly data, we explore methods to improve detection performance. This research investigates the design and evaluation of a novel IDS. We leverage and optimize supervised ML methods like tree-based Support Vector Machines (SVM), ensemble methods, and neural networks (NN) alongside the cutting-edge DL approach of long short-term memory (LSTM) and vision transformers (ViT). We optimized the hyperparameters of these algorithms using a robust Bayesian optimization approach. The implemented ML models achieved impressive training accuracy, with Random Forest and Ensemble Bagged Tree surpassing 99.90% of accuracy, an AUC of 1.00, an F1-score, and a balanced Matthews Correlation Coefficient (MCC) of 99.78%. While the initial deep learning LSTM model yielded an accuracy of 99.97%, the proposed ViT architecture significantly boosted performance with 100% of all metrics, along with a validation accuracy of 78.70% and perfect training accuracy. This study demonstrates the power of our new methods for detecting and stopping attacks on Internet of Things (IoT) networks. This improved detection offers
This study is an optimized extension based on the authors’previous research on the tribo-chemical reaction under constant temperature field of two-stroke internal combustion engines(ICEs).It establishes a coupled ana...
详细信息
This study is an optimized extension based on the authors’previous research on the tribo-chemical reaction under constant temperature field of two-stroke internal combustion engines(ICEs).It establishes a coupled analysis model that considers the tribo-chemical reactions,dynamic contact,and interface lubrication of the piston ring-cylinder liner(PRCL)system under transient temperature *** this study,for the first time,the prediction of the tribofilm thickness and its influence on the surface micro-topography(the comprehensive roughness)are coupled in the working temperature field of the PRCL system,forming an effective model framework and providing a model basis and analytical basis for subsequent *** study findings reveal that by incorporating temperature and tribofilm into the simulation model,the average friction deviation throughout the stroke decreases from 8.92%to 0.93%when compared to experimental ***,the deviation during the combustion regime reduces from 39.56%to 7.34%.The proposed coupled model provides a valuable tool for the evaluation of lubrication performance of the PRCL system and supports the analysis software forward design in two-stroke ICEs.
Accurate reconstruction of particle images is imperative in high-energy physics experiments for the detection and identification of new particles. This paper will introduce how to build a model which can use the data ...
详细信息
The micro-morphology and molecular stacking play a key role in determining the charge transport process and nonradiative energy loss, thus impacting the performances of organic solar cells(OSCs). To address this issue...
详细信息
The micro-morphology and molecular stacking play a key role in determining the charge transport process and nonradiative energy loss, thus impacting the performances of organic solar cells(OSCs). To address this issue, a non-fullerene acceptor PhC6-IC-F with alkylbenzene side-chain, possessing optimized molecular stacking, complementary absorption spectra and forming a cascade energy level alignment in the PM6:BTP-eC9 blend, is introduced as guest acceptor to improve efficiency of ternary OSCs. The bulky phenyl in the side-chain can regulate crystallinity and optimizing phase separation between receptors in ternary blend films, resulting in the optimal phase separations in the ternary films. As a result, high efficiencies of 18.33% as photovoltaic layer are obtained for PhC6-IC-F-based ternary devices with excellent fill factor(FF) of 78.92%. Impressively, the ternary system produces a significantly improved open circuit voltage(V_(oc)) of 0.857 V compared with the binary device,contributing to the reduced density of trap states and suppressed non-radiative recombination result in lower energy loss. This work demonstrates an effective approach for adjusting the aggregation, molecular packing and fine phase separation morphology to increase V_(oc) and FF, paving the way toward high-efficiency OSCs.
For Ultra Reliable Low Latency Communication (URLLC), network delay plays a crucial role in various communication services. However, current temporal models do not perform as remarkably as expected in various network ...
详细信息
ISBN:
(数字)9798350343199
ISBN:
(纸本)9798350343205
For Ultra Reliable Low Latency Communication (URLLC), network delay plays a crucial role in various communication services. However, current temporal models do not perform as remarkably as expected in various network time delay downstream tasks, as their inadequately representation of temporal periodicity and statistical features. This paper proposes a contrastive learning method, ReDL, which derives temporal characteristics from network time delay, to generate notable representation suitable for different tasks. At first, we divide time delay into overlapping temporal sequences as positive samples, so as to effectively preserve temporal continuity. Based on that, we present a novel temporal representation learning method with two innovational steps. On one aspect, a multiscale extractor is employed to obtain temporal features from both the entire time interval and the local ones. On the other aspect, a hierarchical contrastive loss is adopted to update the extractor from temporal and instance dimensions, aiming to accomplish representation in different levels of granularity. The experimental results demonstrate that ReDL can achieve significant temporal representation on datasets in real URLLC scenarios. In time delay prediction, MSE of ReDL is 14.8% less than TS2Vec and 53.1% less than Informer. In anomaly detection, accuracy of ReDL is 5.8% higher than TS-TCC and 7.23% higher than TNC.
Speech emotion recognition is a difficult task that is gaining attention in a variety of domains, including psychology, human–computer interaction, and speech processing. To recognize speech emotions, machine learnin...
详细信息
Data compression plays a vital role in datamanagement and information theory by reducing ***,it lacks built-in security features such as secret keys or password-based access control,leaving sensitive data vulnerable t...
详细信息
Data compression plays a vital role in datamanagement and information theory by reducing ***,it lacks built-in security features such as secret keys or password-based access control,leaving sensitive data vulnerable to unauthorized access and *** the exponential growth of digital data,robust security measures are *** encryption,a widely used approach,ensures data confidentiality by making it unreadable and unalterable through secret key *** their individual benefits,both require significant computational ***,performing them separately for the same data increases complexity and processing *** the need for integrated approaches that balance compression ratios and security levels,this research proposes an integrated data compression and encryption algorithm,named IDCE,for enhanced security and *** on 128-bit block sizes and a 256-bit secret key *** combines Huffman coding for compression and a Tent map for ***,an iterative Arnold cat map further enhances cryptographic confusion *** analysis validates the effectiveness of the proposed algorithm,showcasing competitive performance in terms of compression ratio,security,and overall efficiency when compared to prior algorithms in the field.
Aiming at the problems of complex parameter design and slow system dynamic response in the traditional double closed-loop PI control system of the Vienna rectifier, this paper proposes a novel double closed-loop contr...
详细信息
暂无评论