Search

Data and privacy laws, such as the GDPR, require mobile apps that collect and process the personal data of their citizens to have a legally-compliant policy. Since these mobile apps are hosted on app distribution platforms such as Google Play Store and App Store, the app publishers also require the app developers who wish to submit a new app or make changes to an existing app to be transparent about their app privacy practices regarding handling sensitive user data that requires sensitive permissions such as calendar, camera, microphone. To verify compliance with privacy regulators and app distribution platforms, the app privacy policies and permissions are investigated for consistency. However, little has been done to investigate GDPR completeness checking within the Android permission ecosystem. In this paper, we investigate the design and runtime approaches towards completeness checking of sensitive (‘dangerous’) Android permission policy declarations against GDPR. In this paper, we investigate the design and runtime approaches towards completeness checking of dangerous Android permission policy declarations against GDPR. Leveraging the MPP-270 annotated corpus that describes permission declarations in application privacy policies, six natural language processing and language modelling algorithms are developed to measure permission completeness during runtime while a proof of concept Class Unified Modeling Language Diagram (UML) tool is developed to generate GDPR-compliant permission policy declarations using UML diagrams during design time. This paper makes a significant contribution to the identification of appropriate permission policy declaration methodologies that a developer can use to target particular GDPR laws, increasing GDPR compliance by 12% in cases during runtime using BERT word embedding, measuring GDPR compliance in permission policy sentences, and a UML-driven tool to generate compliant permission declarations.

Link prediction in complex networks is a critical process aimed at uncovering hidden or potential connections among nodes. This technique is widely utilized in areas such as knowledge graphs. Current Graph Neural Networks (GNNs) often focus exclusively on determining whether nodes are connected or assessing the strength of these links by leveraging node attributes. They typically use network structure and attributes to develop node representations through neighborhood aggregation. However, these methods often overlook the intrinsic importance of the links themselves. This paper thoroughly examines the significance of link value based on network structure and introduces an innovative approach for estimating this value, and proposes a method that incorporates link value into both the formulation and training of a link prediction graph attention network. This integration not only boosts the accuracy of link predictions but also provides a theoretical basis for understanding the prediction results. We conducted extensive experiments in link prediction employing widely recognized benchmark datasets. The findings reveal that our proposed framework for link prediction exhibits commendable performance and generalization capabilities, and overall performance improved by an average of 1.2%, thereby establishing it as an effective baseline model.

The limited size of existing datasets and signal variability have hindered EEG-based emotion recognition. In this paper, we present a solution that simultaneously addresses both problems. Generative Adversarial Networks (GANs) have recently shown notable data augmentation (DA) success. Therefore, we leverage a GAN-based DA technique to enhance the robustness of our proposed emotion recognition model by synthetically increasing the size of our datasets. Moreover, we employ contrastive learning to improve the quality of the learned representations from EEG signals and mitigate the adverse impact of inter-subject and intra-subject variability in signals corresponding to the same stimuli or emotions. We do so by maximizing the similarity in the representation of such EEG signals. We perform EEG-based emotion classification using a Graph Neural Network (GNN), which learns the relationship between the extracted EEG features. We compare the proposed model with several recent state-of-the-art emotion recognition models on the DEAP and MAHNOB datasets. The experimental results demonstrate that the proposed model outperforms previous models with a 64.84% and 66.40% emotion classification accuracy on the test set of the DEAP dataset and a 66.98% and 71.69% emotion classification accuracy on the test set of the MAHNOB-HCI dataset for the valence and arousal emotional dimensions, respectively. We perform an ablation study to demonstrate how contrastive learning, GAN, and GNN contribute to improving the proposed solution’s performance.

Cognitive agents make systems autonomous through the process of decision automation by mining an existing knowledge repository at run time. These processes can often be highly compute intensive, and would thus run slowly on the low-power computing platforms typically seen in autonomous systems. This paper examines how knowledge be represented in a Q-table and proposes a novel fast algorithm to mine that knowledge based on constraints. We evaluate this approach for the knowledge mining process of a specific agent: Cognitively Enhanced Complex Event Processing (CECEP). Within CECEP, knowledge is represented using Cognitive Domain Ontologies (CDO), and is mined using situational inputs and constraints. This is a novel approach to store information and is able to accommodate CDOs with millions of solutions. To show that the approach can run on low power hardware in real-time, this algorithm was executed on two low-power minicomputing platforms - Intel’s NUC and Asus’s Tinker Board. At present, no other optimized CDO solvers can generate solutions on these platforms. The algorithm generated the same amount of solutions as a GPU-enabled optimized path-based forward checking CDO solver, while consuming around 7.7 and 5.15 times less energy (Joules) on the NUC and Tinker Board respectively.

The Partial Element Equivalent Circuit (PEEC) method has gained significant recognition as an electromagnetic computational technique known for its ability to represent electromagnetic phenomena using equivalent circuits. This feature makes it particularly valuable for addressing mixed EM-circuit problems. However, PEEC models often exhibit large dimensions, necessitating modeling techniques that can effectively reduce their size while preserving accuracy. Model order reduction (MOR) serves as a highly effective approach to accomplish this objective. This paper presents two MOR techniques based on proper orthogonal decomposition (POD) for PEEC models described by neutral delayed differential equations (NDDEs). The unique characteristics of NDDEs demand specialized MOR approaches, as their formulation is inherently more complex compared to standard quasi-static PEEC models described by non-delayed differential equations. In addition to a traditional one-shot singular value decomposition (SVD), this paper also presents an incrementally computed SVD to evaluate the orthogonal matrix needed to generate the reduced order matrices. The accuracy and efficiency of the proposed PEEC-MOR methods are demonstrated through multiple relevant numerical results in both the frequency-domain and time-domain.

With the development of cyber technology, the intelligence of power systems has increased, and cyber-physics is highly integrated and mutually constrained. In this paper, a globalized cyber-physical cooperative recovery strategy based on node importance is proposed for the failure of a cyber-physical power system under communication failure. First, the static importance of communication nodes and the importance of services are comprehensively analyzed based on the structural business transmission function characteristics of the network. Secondly, based on the fault characteristics of generators, power lines, and loads, the recovery model of the physical system is constructed by considering the safety constraints such as system frequency, node voltage, and line capacity. Finally, considering the possible delay caused by the cyber system failure to the generator output regulation, power line commissioning, and load recovery, the interaction model of physical system recovery and cyber system recovery is established and solved by using the solver after linearizing it. The results show that the cyber-physical recovery strategy can effectively reduce outage losses and improve recovery efficiency.

The goal of music genre classification is to identify the genre of given feature vectors representing certain characteristics of music clips. In addition, to improve the accuracy of music genre classification, considerable research has been conducted on extracting spectral features, which contain critical information for genre classification, from music clips and feeding these features into training models. In particular, recent studies argue that classification accuracy can be enhanced by employing multiple spectral features simultaneously. Consequently, fusing information from multiple spectral features is a critical consideration in designing music genre classification models. Hence, this paper provides a short survey of recent studies on music genre classification and compares the performance of the most recent CNN-based models with a newly devised model that employs a late fusion strategy for the multiple spectral features. Our empirical study of 12 public datasets, including Ballroom, ISMIR04, and GTZAN, showed that the late fusion CNN model outperforms other compared methods. Additionally, we performed an in-depth analysis to validate the effectiveness of the late fusion strategy in music genre classification.

Text-conditioned outfit recommendation aims to recommend a whole fashion outfit that satisfies the compatibility between the recommended items and given items and adheres to the text condition like “Paradise Tropical Vacation” or “60s Style”. Using text description as a condition can provide users with a flexible and accurate way to retrieve and recommend fashion items but this problem is underexplored by existing studies. A challenge of text-conditioned outfit recommendation is how to encode and fuse the outfit text description and fashion item images and text. To solve this, this paper proposes a framework for this task which features a hybrid attention layer that constructs the relationship between outfit text description and fashion items for condition compliance, and the relationship between fashion items for internal compatibility. To encode fashion item features, our method uses pre-trained FashionCLIP as an extractor which significantly reduces the trainable parameters compared to previous methods training CNN from scratch. The whole outfits are generated by iteratively adding compatible items based on a given partial outfit. Compared with state-of-the-art methods on polyvore disjoint and non-disjoint datasets, our approach can achieve 3% relative improvement in compatibility prediction AUC, achieve 5% relative improvement in fill-in-the-blank accuracy; achieve 19% relative improvement on complementary item retrieval recall at different ranks in average. Besides, We demonstrate that our approach can recommend a whole outfit with inner compatibility and adhere to the text description.

In a rapidly advancing world of technology, information security studies have become the backbone of the digital age, and steps in this area are critical. In this context, cryptography, in particular, plays a key role in ensuring the confidentiality, integrity and authentication of data. s-box structures provide a certain diversity and security layer in encryption algorithms, forming one of the key elements in this area. This study focuses on the design and analysis of s-box structures, examining the potential impact of chaos theory-based structures on encryption systems. First, it provides a comprehensive classification of existing s-box design proposals in the literature, and explores the contribution of chaos theory to the security features of these structures. The original contribution of the study is the results obtained with the help of the developed analysis and design program. The program optimizes levels of complexity, randomness, and resistance, and demonstrates the resistance of these new structures to cryptanalysis attacks. The paper also draws attention to open issues in the field of chaos-based s-box design and provides a road map for future research. It is estimated that all these findings will provide a common motivation for researchers in the relevant literature and constitute the basis for many practical practices.

A key feature of 5G systems is the Ultra-Reliable Low-Latency Communication (URLLC), which can be used for remote surgery, smart grids, industrial control, etc. URLLC requires millisecond-level delays and very high reliability, i.e., less than $10^{-5}$ packet loss probability. The ability to satisfy these very strict quality of service requirements depends on selecting the Modulation and Coding Schemes (MCS) for data transmissions. On the one hand, the selected MCS shall be robust enough to avoid multiple retransmissions within a small delay budget. On the other hand, the MCS shall be high-rate to reduce channel resource consumption and, thus, shall increase the system capacity for URLLC. The MCS selection problem is extremely challenging to capture the quickly varying wireless channel effects, e.g., in highly mobile scenarios, because the decision shall be made long before the actual transmission occurs. The paper proposes a novel MCS selection algorithm called ALPACA (Asymmetric Loss Prediction Algorithm for Channel Adaptation), which relies on a widely used class of convolutional-recurrent neural networks. However, in contrast to existing approaches, ALPACA explicitly considers the asymmetric error cost for channel prediction by utilizing quantile regression loss. Both real-life channel measurements and 3GPP channel models are used to evaluate the performance of ALPACA. Numerical results demonstrate the increase in the reliability and reduction in resource consumption compared with the existing MCS selection algorithms, which results in 40% growth of the network capacity.

Learning high-quality representations of users, items, and tags from historical interactive data is crucial for personalized tag recommendation (PTR) systems. Currently, most PTR models are committed to learning representations from first-order interactions without considering the exploitation of high-order interactive relations, which can be beneficial for avoiding sub-optimal learning. Although several PTR models equipped with graph neural networks (GNN) have been proposed to capture higher-order semantic relevance from raw data, they all carry out representation learning in Euclidean space, which can still easily result in sub-optimal learning due to embedding distortion. In order to further improve the quality of representation learning for PTR, the paper proposes a novel PTR model based on a lightweight GNN framework with hyperbolic embedding, namely GHPTR. GHPTR explicitly injects higher-order relevance into entity representation through the message propagation and aggregation mechanism of GNN and leverages hyperbolic embedding to alleviate the embedding distortion problem. Experimental results on real-world datasets have demonstrated the superiority of our model over its Euclidean counterparts and state-of-the-art baselines.

This paper addresses the joint design of a beamformer and user-centric clustering for scalable cell-free massive multiple-input multiple-output (CF-mMIMO) under severe quantization noise generated in backhaul links for central processing units (CPUs) cooperation. The system model comes up with a plan in which multiple CPUs exchange physical layer data under limited bandwidth to enhance performance while introducing clustering across multiple CPUs. We derive the joint optimization of the minimum mean-square error (MMSE) beamformer and user-centric clustering under quantization noise, and propose its low-complexity design. The superiority of our proposed design method is clarified by comparing its performance with cellular distributed MIMO systems. Throughout the paper, we first answer the problem of how much gain CF-mMIMO systems with multiple CPUs cooperation can obtain.

Heterogeneous cross-project defect prediction (HCPDP) aims to predict defects in new software projects using defect data from previous software projects where the source and target projects have some different metrics. Most existing methods only find linear relationships in the software defect datasets. Additionally, these methods use multiple defect datasets from different projects as source datasets. In this paper, we propose a novel method called heterogeneous cross-project defect prediction using encoder networks and transfer learning (ENTL). ENTL uses encoder networks to extract the important features from source and target datasets. Also, to minimize the negative transfer during transfer learning, we used an augmented dataset that contains pseudo-labels and the source dataset. Additionally, we have used a single dataset to train the model. To evaluate the performance of the ENTL approach, 16 datasets from four publicly available software defect projects were used. Furthermore, we compared the proposed method with four HCPDP methods namely EGW, HDP_KS, CTKCCA and EMKCA, and one WPDP method from existing literature. The proposed method on average outperforms the baseline methods in terms of PD, PF, F1-score, G-mean and AUC.

The high-speed vibration rotation of the drill bit during drilling causes the logging tool to be damaged or distorted, resulting in inaccurate or lost data collection. Traditional prediction methods such as dynamic modeling and geological modeling have problems such as incomplete data and difficult modeling, which cannot meet the accuracy and stability requirements of wellbore trajectory prediction. The long short-term memory neural network (LSTM) for predicting time series can achieve accurate prediction, but there are problems such as difficulty in adjusting the hyperparameters of the LSTM model, slow convergence speed, and easy overfitting. This paper absorbs the advantages of the LSTM algorithm, ridge regression (L2 regularization), and sparrow optimization algorithm (SSA) in machine learning and proposes a well trajectory prediction model of steerable drilling based on L2 regularization and SSA optimized LSTM (L2-SSA-LSTM). The model takes the LSTM hyperparameter as the parameter optimization goal of SSA and adds L2 regularization to the model to prevent model overfitting to complete modeling and prediction. The experiment was conducted using measured data sets from directional drilling in two different oilfields. The results show that compared with the back propagation algorithm (BP), consolidated memory gated recurrent unit (CMGRU), dual-thread gated recurrent unit (DTGRU), Attention-based Spatiotemporal Graph Recurrent Neural Network (ASTG-RNN), LSTM, and the L2-SSA-LSTM prediction model has significantly higher accuracy in predicting directional drilling trajectories than other models and has the better predictive ability.

Multi-channel FinFET ( $\text{M}_{\textbf {ch}}$ -FinFET) is an emerging device having promising use due to its excellent driving capability. In this paper, we have investigated the significance of multiple channels of FinFET configuration. We have examined the performance of the multi-channel-based step FinFET ( $\text{M}_{\textbf {ch}}$ - step FinFET) structure with spacer engineering. The results obtained from this simulation work indicate that $\text{M}_{\textbf {ch}}$ -step FinFET is a good competitor for future improvisation of CMOS technology. The proposed device has improved drain conductivity, transconductance ( ${\textbf G_{m}}$ ), intrinsic gain ( ${\textbf A_{v}}$ ), and drain conductance ( ${\textbf G_{d}}$ ) performance by introducing high-K dielectric spacer material by 38.2%, 46.12%, 88.57%, and 22.55%, respectively. The proposed device with a spacer is preferable to obtain better performance regarding ON current and device efficiency.

The fifth-generation (5G) wireless communications have been deployed in many countries with the following features: wireless networks at 20 Gbps as peak data rate, a latency of 1-ms, reliability of 99.999%, maximum mobility of 500 km/h, a bandwidth of 1-GHz, and a capacity of 106 up to Mbps/m2. Nonetheless, the rapid growth of applications, such as extended/virtual reality (XR/VR), online gaming, telemedicine, cloud computing, smart cities, the Internet of Everything (IoE), and others, demand lower latency, higher data rates, ubiquitous coverage, and better reliability. These higher requirements are the main problems that have challenged 5G while concurrently encouraging researchers and practitioners to introduce viable solutions. In this review paper, the sixth-generation (6G) technology could solve the 5G limitations, achieve higher requirements, and support future applications. The integration of multiple access techniques, terahertz (THz), visible light communications (VLC), ultra-massive multiple-input multiple-output ( $\mu {\mathrm{ m}}$ -MIMO), hybrid networks, cell-free massive MIMO, and artificial intelligence (AI)/machine learning (ML) have been proposed for 6G. The main contributions of this paper are a comprehensive review of the 6G vision, KPIs (key performance indicators), and advanced potential technologies proposed with operation principles. Besides, this paper reviewed multiple access and modulation techniques, concentrating on Filter-Bank Multicarrier (FBMC) as a potential technology for 6G. This paper ends by discussing potential applications with challenges and lessons identified from prior studies to pave the path for future research.

Next generation of wireless networks are characterized by two main features named Enhanced Mobile Broadband (eMBB) and Ultra Reliable Low Latency Communications (URLLC). These two services can be accommodated in the same wireless infrastructure so that wide range of users, demanding either massive throughput or extremely low latency and high reliability requirements, are directly benefited for providing various mission critical services. Co-existence of eMBB and URLLC services, however, demand highly efficient and less complex resource allocation schemes. In this paper, various resource allocation techniques are studied for the co-existence of eMBB and URLLC traffic to meet the heterogeneous specifications of each class of users. A detailed study on existing resource allocation schemes for simultaneous transmission of eMBB and URLLC services based on network slicing, flexible Transmit Time Interval (TTI), scheduling and distributed and federated learning are provided. Moreover, Machine Learning (ML) aided and Reconfigurable Intelligent Surface (RIS) and Unmanned Aerial Vehicle (UAV) assisted resource allocation techniques are also studied in detail. Additionally, this paper identifies some challenges for eMBB and URLLC service accommodations in the same wireless architecture, and proposes their possible solution approaches.

Federated learning (FL) is now considered a critical method for breaking down data silos. However, data encryption can significantly increase computing time, limiting its large-scale deployment. While hardware acceleration can be an effective solution, existing research has largely focused on a single hardware type, which hinders the acceleration of FL across the various heterogeneous hardware of the participants. In light of this challenge, this paper proposes a novel FL acceleration framework that supports diverse types of hardware. Firstly, we conduct an analysis of the key elements of FL to clarify our accelerator design goals. Secondly, a unified acceleration framework is proposed, which divides FL into four layers, providing a basis for the compatibility and implementation of heterogeneous hardware acceleration. After that, based on the physical properties of three mainstream acceleration hardware, i.e., GPU, ASIC and FPGA, the architecture design of corresponding heterogeneous accelerators under the framework is detailed. Finally, we validate the effectiveness of the proposed heterogeneous hardware acceleration framework through experiments. For specific algorithms, our implementation achieves a state of the art acceleration effect compared to previous work. For the end-to-end acceleration performance, we gain $12\times $ , $7.7\times $ and $2.2\times $ improvement on GPU, ASIC and FPGA respectively, compared to CPU in large-scale vertical linear regression training tasks.

Brain-computer interface (BCI) systems have been developed to assist individuals with neuromuscular disorders to communicate with their surroundings using their brain signals. One attractive branch of BCI is steady-state visual evoked potential (SSVEP), which has acceptable speed and accuracy and is non-invasive. However, SSVEP-based EEG signals suffer from eye-fatigue problems, resulting in artifacts that affect the accuracy of the system. Thus, researchers are still working to improve SSVEP-based BCI systems. This paper proposes robust machine-learning algorithm for single-flicker SSVEP detection. A novel approach based on fast independent component analysis and filter-bank canonical correlation analysis (fast ICA-FBCCA) is developed to extract features from the single-flicker SSVEP signal. The clean features learned by fast ICA-FBCCA are then applied to a discrete wavelet transform (DWT) technique and fed to a convolutional neural network (CNN) with only one convolutional layer and a smaller number of parameters. The effectiveness of the proposed technique is evaluated using two datasets. The results were evaluated using two datasets. The findings clearly demonstrate that the proposed method outperforms traditional methods, with average target recognition accuracy and standard deviation values of 97 ± 3.1% among 6 subjects for dataset 1 and 82.12 ± 10.7% among 12 subjects for dataset 2. Overall, these findings suggest that the proposed method is a promising approach for improving the accuracy and reliability of the single-flicker SSVEP-based BCI systems.

Advanced Driver Assistance Systems (ADAS) are advanced technologies that assist drivers with vehicle operation and navigation. Recent improvements and brisk expansion in the ADAS market, as well as an increase in the frequency of incidents such as sensor spoofing, communication interruption etc., in autonomous vehicles (AVs), have raised the need to research ADAS security technology. The security issues raised by incorporating these technologies into automobiles must be addressed to protect the privacy and safety of passengers and other road users. As a result, the purpose of this research is to investigate the security issues that arise from the integration of ADAS technologies. Addressing these challenges holds the potential to establish a foundation for enhanced safety and dependability within transportation networks amidst the ongoing advancements in vehicle technology. This paper starts by describing the vulnerabilities, threats, assaults, and defense mechanisms of the ADAS. It then delves into the attacks and countermeasures in three categories, namely VANET, Hardware, and Adversarial attacks. VANET attacks encompass threats targeting Vehicular Ad Hoc Networks, aiming to disrupt communication among vehicles or between vehicles and infrastructure. Hardware attacks focus on vulnerabilities within the physical components of ADAS, including sensors, processors, or communication modules. Adversarial attacks involve deliberate manipulations or perturbations introduced into machine learning models or algorithms utilized within ADAS. These attacks aim to deceive or undermine the functionality of AI-based systems, causing misclassification, compromising system integrity, and posing risks to user safety by exploiting vulnerabilities in the AI decision-making process. Finally, this study highlights potential areas for future research, such as the utilization of artificial intelligence (AI), the necessity of industry-wide standardization, and recommends specific future work tailored to each attack described in the corresponding sections.