IEEE Access

Journal Information
EISSN: 21693536
Total articles ≅ 65,332

Latest articles in this journal

Muhammad Adnan, , Emel Khan, , Samina Amin, Ahmad A. Alzahrani
In this research study, we propose an Explainable Artificial Intelligence (XAI) model that provides the earliest possible global and local interpretation of students’ performance at various stages of course length. Global and local interpretation is provided in such a way that the prediction accuracy of a single local observation is close to the model’s overall prediction accuracy. For the earliest possible understanding of student performance, local and global interpretation is provided at 20%, 40%, 60%, 80%, and 100% of course length. Machine Learning (ML) and Deep Learning (DL) which are subfields of Artificial Intelligence (AI) have recently emerged to assist all educational institution’s in predicting the performance, engagement, and dropout rate of online students. Unfortunately, traditional ML and DL techniques lack in providing data analysis results in an understandable human way. Explainable AI (XAI), a new branch of AI, can be used in educational settings, specifically in VLEs, to provide the instructor with the study performance results of thousands or even millions of online students in a human-understandable way. Thus, unlike black box approaches such as traditional ML and DL techniques, XAI can help instructors to interpret the strengths and weaknesses of an individual student, providing them with timely personalized feedback and guidance. Various traditional and various ensemble ML algorithms were trained on demographic, clickstream, and assessment features to determine which algorithm gives the best performance result. The best-performing ML algorithm was ultimately selected and provided to the XAI model as an input for local and global interpretation of students’ study behavior at various percentages of course length. We have used various XAI tools to give students’ performance reports to instructors, in an explicable human way, at different stages of course length. The intermediate data analysis and performance reports will help instructors and all key stakeholders in decision-making and optimally facilitate online students.
Ju-Young Lee, Hyo-Seo Jang, Jeong-Il Kang,
The conventional power factor correction (CPFC) converter suffers from excessive heat generation due to high conduction loss from the input bridge diode. Various bridgeless PFCs (BPFCs) have been proposed to overcome this drawback. Among them, the conventional BPFC (CBPFC) features the significantly reduced conduction loss but has very poor conducted electromagnetic interference (EMI) noise. Further, the semi-BPFC features the reduced input diode loss and good EMI, but the input diode loss is too significant to ignore. Although another solution, totem-pole BPFC, features high efficiency and good EMI, it usually requires expensive current sensors and microprocessor. A BPFC with the integrated magnetics common mode coupled inductor (ICC) is proposed to overcome the drawbacks of traditional BPFCs. The ICC can significantly reduce the system size by integrating three inductive components into a single magnetic core. Moreover, the small magnetizing current of the ICC, not the main input current, flows through input diodes; thus, experimental results from a 600 W rated prototype prove that the input diode loss can be significantly reduced by about 10 W compared to the CPFC, and the input diode exhibits a low heat generation of 49.8 °C despite the absence of a heat sink. In particular, with the help of input diodes, the EMI noise spectra of the ICC BPFC can meet the EMI standard with a sufficient margin. The detailed analysis, design guide and experimental results from a 600 W rated prototype are provided to confirm the validity of the ICC BPFC.
, , , Karim H. Moussa, , , Norah Muhammad Alwadai, Heba G. Mohamed
In this manuscript, a multiple-input multiple-output (MIMO) antenna array system with identical compact antenna elements providing wide radiation and diversity function is introduced for sub 6 GHz fifth-generation (5G) cellular applications. The introduced design contains four pairs of miniaturized square-loop resonators with dual-polarization and independently coupled T-shaped feed lines which have been placed symmetrically at the edge corners of the smartphone mainboard with an overall size of 75 mm × 150 mm. Therefore, in total, the introduced array design encompasses four pairs of horizontally and vertically polarized resonators. The elements are very compact and utilize at 3.6 GHz, a potential 5G candidate band. In order to improve the frequency bandwidth and radiation coverage, a square slot has been placed and excited under each loop resonator. Desirable isolation has been observed for the adjacent elements without any decoupling structures. Therefore, they can be considered self-isolated elements. The presented smartphone antenna not only exhibits desirable radiation but also supports different polarizations at various sides of the printed circuit board (PCB). It exhibits good bandwidth of 400 MHz (3.4-3.8 GHz), high-gain patterns, improved radiation coverage, and low ECC/TARC (better than 0.004 and -30 dB at 3.6 GHz, respectively). Experimental measurements were conducted on an array manufactured on a standard smartphone board. The simulated properties of this MIMO array are compared with the measurements, and it is found that they are in good agreement. Furthermore, the introduced smartphone array offers adequate efficiency in both the user interface and components integrated into the device. As a result, it could be suitable for 5G handheld devices.
Xuan Liu, Chenfeng Zhang, Yingzhi Wang, Kai Ding, , Hong Liu, Yu Tian, Bo Xu, Mingchi Ju
At present, researchers have made great progress in the research of object detection, however, these studies mainly focus on the object detection of images under normal lighting, ignoring the target detection under low light. And images in the fields of automatic driving at night and surveillance are usually obtained in low-light environments. These images have problems such as poor brightness, low contrast, and obvious noise, which lead to a large amount of information loss in the image. And the performance of object detection in low light is reduced. In this paper, we propose a low-light image enhancement method based on multi-scale network fusion to solve the problems of images in low-light environments. Aiming at the problem that the effective information of low-light images is relatively small, we propose a preprocessing method for image nonlinear transformation and fusion, which improves the amount of available information in the light image. Then, in order to obtain a better enhancement effect, a multi-scale feature fusion method is proposed, which fuses features from different resolution levels in the network. The details of low-light areas in the image are improved, and the problem of feature loss caused by too deep network layers is solved. The experimental results show that our proposed method can achieve better enhancement effects on different datasets compared with the current mainstream methods. The average recall value of the object detection with our method is improved by 38.25%, which shows that our proposed method is effective and can promote the development of autonomous driving, monitoring, and other fields.
Junhyuk Hyun, Suhan Woo,
In urban cities, the information about the type or class of street floors enables a wheeled mobile robot to perform many tasks ranging from traversability region identification, localization and the choice of wheel control strategy. In this paper, we considered a new task named as street floor segmentation (SFS) using an RGB camera. The SFS can be considered as the generalized problem of the existing traversability region identification problem in urban situations. Our SFS has two special classes for the possible application to the traversability region identification and they are traversable and non-traversable curbs. The SFS using an RGB camera is implemented using a real-time semantic segmentation (SS) network. A booster module named as Dynamic Context-based Refinement Module (DCRM) was developed to enhance the performance of the SFS. Our network was applied to real-world applications, and its validity is demonstrated through experiment.
Zhenchao Cui, Ziang Chen, , Zhaoqi Wang
Sign language production aims to automatically generate coordinated sign language videos from spoken language. As a typical sequence to sequence task, the existing methods are mostly to regard the skeletons as a whole sequence, however, those do not take the rich graph information among both joints and edges into consideration. In this paper, we propose a novel method named Spatial-Temporal Graph Transformer (STGT) to deal with this problem. Specifically, according to kinesiology, we first design a novel graph representation to achieve graph features from skeletons. Then the spatial-temporal graph self-attention utilizes graph topology to capture the intra-frame and inter-frame correlations, respectively. Our key innovation is that the attention maps are calculated on both spatial and temporal dimensions in turn, meanwhile, graph convolution is used to strengthen the short-term features of skeletal structure. Finally, due to the generated skeletons are based on the form of skeleton points and lines so far. In order to visualize the generated sign language videos, we design a sign mesh regression module to render the skeletons into skinned animations including body and hands posture. Comparing with states of art baseline on RWTH-PHONEIX Weather-2014T in Experiment Section, STGT can obtain the highest values on BLEU and ROUGE, which indicates our method produces most accurate and intuitive sign language videos.
A compact, dual polarized, multiband four-port flexible Multiple Input Multiple Output (MIMO) antennae with the connected ground and high isolation is designed with computation and experimental measurement studies. All four monopole radiators are embedded decagon-shaped flexible FR-4 substrate with an outer radius of 10 mm in order to accomplish circularly polarized (CP) radiations, bandwidth enhancement, and compact size of only 45×38×0.2 mm 3 (0.375λ × 0.316λ × 0.0016λ, at lowest resonating frequency 2.5GHz). The interconnected ground structure is loaded with an Interlaced Lozenge Structure (ILS) to suppress the surface wave radiations resulting in low mutual coupling between the radiators. The proposed MIMO antenna demonstrates measured 10-dB impedance bandwidths of 9.63% (2.37–2.61 GHz), 28.79% (3.30–4.41 GHz), and 16.91% (4.98–5.90 GHz) in the LTE 38/40, Sub-6 GHz 5G NR n77/n78, WLAN and Wi-Fi bands, respectively. Furthermore, broad 3-dB Axial Ratio Bandwidth (ARBW) of 28.79% (3.30–4.41 GHz) with gain greater than 4 dBi and efficiency above 80% are achieved. Finally, the bending analysis of the proposed flexible MIMO antenna along the X- and Y- directions shows good performances in terms of scattering parameters, 3 dB ARBW, and MIMO diversity parameters.
Jie Wu, Li Wang,
This paper mainly addresses bipartite containment control of multi-agent systems (MASs) subject to exogenous disturbances. With dynamic gain technique independent of any global information, the disturbance observer method is applied to estimate disturbances generated by heterogeneous nonlinear exosystems. Two disturbance observer-based controllers are accordingly presented via state feedback approach and output feedback approach. By means of appropriate Lyapunov method, it is shown that the bipartite containment control is realized under sufficient criteria. Finally, simulations are employed to validate the effectiveness and correctness of our proposed controllers.
Alba Vadillo-Valderrama, , Raul Caulier-Cisterna, Arcadi Garcia-Alberola,
Holter systems record the electrocardiogram (ECG), which is used to identify beat families according to their origin and severity. Many systems have been proposed using signal conditioning and machine learning (ML) classification algorithms for beat family recognition. However, the design stage of these systems does not always consider the impact that tuning the intermediate blocks has on the beat family classification and the overall accuracy. We propose to use a new index based on the confusion matrices and bootstrap resampling to summarize the global performance for all family beats, so-called differential beat accuracy (DBA), which is obtained as the total number of beats correctly classified in each class minus the total number of beats incorrectly classified. We addressed the sensitivity of the different subblocks when creating a simple beat family classifier consisting of signal preprocessing blocks and a simple k-Nearest Neighbors classifier. The MIT-BIH Arrhythmia database was used for this purpose, following existing literature on the field.We benchmarked two implementations, one for biclass classification (supraventricular vs. non-supraventricular origin) and another for multiclass beat labeling. The usual preprocessing stages were scrutinized with the DBA to evaluate their impact on the quality of the complete ML system, such as signal detrending and filtering, beat balancing, or inter-beat distance. With the support of the DBA, our methodology was able to detect significant differences in terms of some of the options in the algorithm design. For instance, balancing the number of beats in each class for training significantly improved the classification accuracy of the minority classes at 3.22% for the multiclass dataset but not for the biclass dataset. Also, accuracy improved significantly by about 6% for the biclass regrouping without data normalization, whereas overall accuracy improved significantly by about 7% for the multiclass regrouping with data normalization. In addition, the analysis of the statistical dispersion of confusion matrices showed that this database should be considered with caution when training ML-based family classifiers. We can conclude that the proposed DBA can provide us with statistically principled criteria for designing ML-based classifiers and reducing their bias in strongly unbalanced beat family datasets.
Robin Wydaeghe, , , Gunter Vermeeren, , ,
Realistic human downlink exposure at 3.5 and 28 GHz to electromagnetic fields is evaluated for distributed and collocated base stations using a hybrid ray-tracing/finite-difference time-domain method. For the first time, the absorbed power density is computed for distributed massive multiple-input multiple-output (DMaMIMO) 6G base stations at 28 GHz. The results are compared with 3.5 GHz 5G base stations. Computational costs are drastically increased at 28 GHz. A large analysis is realized by speed improvements and using two configurations. In the first, exposure distributions of DMaMIMO BS show clusters of low and high exposure. These clusters disappear when results are normalized with respect to the incoming power at the user. In the second, the influence of BS to user distance in line-of-sight (LOS) and non-line-of-sight (NLOS) scenarios shows expected results. This includes a power law relationship in LOS and shadowing in NLOS. The vast majority of exposure quantities are less than 4% of the limits of the International Commission for Non-Ionizing Radiation. Basic restrictions are respected when reference quantities are set to their limits. With equal power, distributed base stations contribute 2 to 3 times less to exposure than collocated base stations. Expressed as a ratio to their limits set by ICNIRP, the basic quantities are 5 to 10 dB lower than the reference quantities.
Back to Top Top