Global software engineering education faces unique challenges to reflect as close as possible real-world distributed team development in various forms. The complex nature of planning, collaborating, and upholding partnerships present administrative difficulties on top of budgetary constrains. These lead to limited opportunities for students to gain international experiences and for researchers to propagate educational and practical insights. This paper presents an empirical view on three different course structures conducted by the same research and educational team over a four-year time span. The courses were managed in Japan and Germany, facing cultural challenges, time-zone differences, language barriers, heterogeneous and homogeneous team structures, amongst others. Three semesters were carried out before and one during the Covid-19 pandemic. Implications for a recent focus on online education for software engineering education and future directions are discussed. As administrational and institutional differences typically do not guarantee the same number of students on all sides, distributed teams can be 1. balanced, where the number of students on one side is less than double the other, 2. unbalanced, where the number of students on one side is significantly larger than double the other, or 3. one-sided, where one side lacks students altogether. An approach for each of these three course structures is presented and discussed. Empirical analyses and reoccurring patterns in global software engineering education are reported. In the most recent three global software engineering classes, students were surveyed at the beginning and the end of the semester. The questionnaires ask students to rank how impactful they perceive factors related to global software development such as cultural aspects, team structure, language, and interaction. Results of the shift in mean perception are compared and discussed for each of the three team structures.

In this letter, we propose a novel transferable sparse regression (TSR) method, for cross-database facial expression recognition (FER). In TSR, we firstly present a novel regression function to regress the data into a latent representation space instead of a strict binary label space. To further alleviate the influence of outliers and overfitting, we impose a row sparsity constraint on the regression term. And a pairwise relation term is introduced to guide the feature transfer learning. Secondly, we design a global graph to transfer knowledge, which can well preserve the cross-database manifold structure. Moreover, we introduce a low-rank constraint on the graph regularization term to uncover additional structural information. Finally, several experiments are conducted on three popular facial expression databases, and the results validate that the proposed TSR method is superior to other non-deep and deep transfer learning methods.

We analyze the effect of window choice on the zero-padding method and corrected quadratically interpolated fast Fourier transform using a harmonic signal in noise at both high and low signal-to-noise ratios (SNRs) on a theoretical basis. Then, we validate the theoretical analysis using simulations. The theoretical analysis and simulation results using four traditional window functions show that the optimal window is determined depending on the SNR; the estimation errors are the smallest for the rectangular window at low SNR, the Hamming and Hanning windows at mid SNR, and the Blackman window at high SNR. In addition, we analyze the simulation results using the signal-to-noise floor ratio, which appears to be more effective than the conventional SNR in determining the optimal window.

The Canonical Polyadic Decomposition (CPD) is the tensor analog of the Singular Value Decomposition (SVD) for a matrix and has many data science applications including signal processing and machine learning. For the CPD, the Alternating Least Squares (ALS) algorithm has been used extensively. Although the ALS algorithm is simple, it is sensitive to a noise of a data tensor in the applications. In this paper, we propose a novel strategy to realize the noise suppression for the CPD. The proposed strategy is decomposed into two steps: (Step 1) denoising the given tensor and (Step 2) solving the exact CPD of the denoised tensor. Step 1 can be realized by solving a structured low-rank approximation with the Douglas-Rachford splitting algorithm and then Step 2 can be realized by solving the simultaneous diagonalization of a matrix tuple constructed by the denoised tensor with the DODO method. Numerical experiments show that the proposed algorithm works well even in typical cases where the ALS algorithm suffers from the so-called bottleneck/swamp effect.

Coprime (pair of) DFT filter banks (coprime DFTFB), which process signals like a spectral analyzer in time domain, divides the power spectrum equally into MN bands by employing two DFT filter banks (DFTFBs) of size only M and N respectively, where M and N are coprime integers. With coprime DFTFB, frequencies in wide sense stationary (WSS) signals can be effectively estimated with a much lower sampling rates than the Nyquist rates. However, the imperfection of practical FIR filter and the correlation based detection mode give rise to two kinds of spurious peaks in power spectrum estimation, that greatly limit the application of coprime DFTFB. Through detailed analysis of the spurious peaks, this paper proposes a modified spectral analyzer based on dual coprime DFTFBs and sub-decimation, which not only depresses the spurious peaks, but also improves the frequency estimation accuracy. The mathematical principle proof of the proposed spectral analyzer is also provided. In discussion of simultaneous signals detection, an O-extended MN-band coprime DFTFB (OExt M-N coprime DFTFB) structure is naturally deduced, where M, N, and O are coprime with each other. The original MN-band coprime DFTFB (M-N coprime DFTFB) can be seen a special case of the OExt M-N coprime DFTFB with extending factor O equals ‘1’. In the numerical simulation section, BPSK signals with random carrier frequencies are employed to test the proposed spectral analyzer. The results of detection probability versus SNR curves through 1000 Monte Carlo experiments verify the effectiveness of the proposed spectrum analyzer.

We present a novel circularly polarized ring microstrip antenna and its design. The shorting pins discretely disposed on the inner edge of the ring microstrip antenna are introduced as a new degree of freedom for improving the resonance frequency control. The number and diameter of the shorting pins control the resonance frequency; the resonance frequency can be almost constant with respect to the inner/outer diameter ratio, which expands the use of the ring microstrip antenna. The dual-band antenna where the proposed antenna includes another ring microstrip antenna is designed and measured, and simulated results agree well with the measured one.

As an emerging technology, device-free localization (DFL) using wireless sensor networks to detect targets not carrying any electronic devices, has spawned extensive applications, such as security safeguards and smart homes or hospitals. Previous studies formulate DFL as a classification problem, but there are still some challenges in terms of accuracy and robustness. In this paper, we exploit a generalized thresholding algorithm with parameter p as a penalty function to solve inverse problems with sparsity constraints for DFL. The function applies less bias to the large coefficients and penalizes small coefficients by reducing the value of p. By taking the distinctive capability of the p thresholding function to measure sparsity, the proposed approach can achieve accurate and robust localization performance in challenging environments. Extensive experiments show that the algorithm outperforms current alternatives.

Deep neural networks show good performance in image recognition, speech recognition, and pattern analysis. However, deep neural networks show weaknesses, one of which is vulnerability to backdoor attacks. A backdoor attack performs additional training of the target model on backdoor samples that contain a specific trigger so that normal data without the trigger will be correctly classified by the model, but the backdoor samples with the specific trigger will be incorrectly classified by the model. Various studies on such backdoor attacks have been conducted. However, the existing backdoor attack causes misclassification by one classifier. In certain situations, it may be necessary to carry out a selective backdoor attack on a specific model in an environment with multiple models. In this paper, we propose a multi-model selective backdoor attack method that misleads each model to misclassify samples into a different class according to the position of the trigger. The experiment for this study used MNIST and Fashion-MNIST as datasets and TensorFlow as the machine learning library. The results show that the proposed scheme has a 100% average attack success rate for each model while maintaining 97.1% and 90.9% accuracy on the original samples for MNIST and Fashion-MNIST, respectively.

Active network monitoring based on Boolean network tomography is a promising technique to localize link failures instantly in transport networks. However, the required set of monitoring trails must be recomputed after each link failure has occurred to handle succeeding link failures. Existing heuristic methods cannot compute the required monitoring trails in a sufficiently short time when multiple-link failures must be localized in the whole of large-scale managed networks. This paper proposes an approach for computing the required monitoring trails within an allowable expected period specified beforehand. A random walk-based analysis estimates the number of monitoring trails to be computed in the proposed approach. The estimated number of monitoring trails are computed by a lightweight method that only guarantees partial localization within restricted areas. The lightweight method is repeatedly executed until a successful set of monitoring trails achieving unambiguous localization in the entire managed networks can be obtained. This paper demonstrates that the proposed approach can compute a small number of monitoring trails for localizing all independent dual-link failures in managed networks made up of thousands of links within a given expected short period.

In this paper, we present a classification method for a Computer-Aided Diagnosis (CAD) system in a colorectal magnified Narrow Band Imaging (NBI) endoscopy. In an endoscopic video image, color shift, blurring or reflection of light occurs in a lesion area, which affects the discrimination result by a computer. Therefore, in order to identify lesions with high robustness and stable classification to these images specific to video frame, we implement a CAD system for colorectal endoscopic images with the Convolutional Neural Network (CNN) feature and Support Vector Machine (SVM) classification on the embedded DSP core. To improve the robustness of CAD system, we construct the SVM learned by multiple image sizes data sets so as to adapt to the noise peculiar to the video image. We confirmed that the proposed method achieves higher robustness, stable, and high classification accuracy in the endoscopic video image. The proposed method also can cope with differences in resolution by old and new endoscopes and perform stably with respect to the input endoscopic video image.

In order to satisfy the user's demands for faster data rates and higher channel capacity, fifth generation (5G) wireless networks operate in the frequency at both sub-6GHz and millimeter wave bands for more abundant spectrum resources. Compared with the sub-6G bands, signals transmitted in the millimeter bands suffer from severe channel attenuation. A drone small cell (DSC) has been proposed recently to provide services outdoors. Not only does DSC have high maneuverability, it can also be deployed quickly in the required regions. Therefore, it is an important issue to establish the Air-to-Ground (ATG) channel model by taking into account the effects of building shielding and excess loss in various DSC deployments at different frequency bands. In this paper, we synthesize the ATG channels of the DSC and approximate the excess path loss of the ATG for different urban environments based on the ITU-R standard. With the approximated curve fitting relations, the proper height of the drone base station that satisfies a certain connected probability can be easily obtained for different scenarios.

In this paper, the influence of surface waves on the characteristics of on-glass antennas is clarified to enable appropriates design of C-band automotive on-glass antennas. Composite glasses are used in automotive windshields. These automotive composite glasses are composed of three layers. First, the surface wave properties of composite glass are investigated. Next, the effects of surface waves on the reflection coefficient characteristics of on-glass antennas are investigated. Finally, the antenna placement to reduce surface wave effect will be presented. Electromagnetic field analysis of a dipole antenna placed at the center of a 300mm × 300mm square flat composite glass showed that the electric field strength in the glass had ripples with the half wavelength period of the surface waves. Therefore, it was confirmed that standing waves are generated because of these surface waves. In addition, it is confirmed that ripples occur in the reflection coefficient at frequencies. Glass size is divisible by each of those guide wavelengths. Furthermore, it was clarified that the reflection coefficient fluctuates with respect to the distance between the antenna and a metal frame, which is attached to the end face in the direction perpendicular to the thickness of the glass because of the influence of standing waves caused by the surface waves; additionally, the reflection coefficient gets worse when the distance between the antenna and the metal frame is an integral multiple of one half wavelength. A similar tendency was observed in an electric field analysis using a model that was shaped like the actual windshield shape. Because radiation patterns also change as a result of the influence of surface waves and metal frames, the results imply that it is necessary to consider the actual device size and the metal frames when designing automotive on-glass antennas.

With the arrival of 5G and the popularity of smart devices, indoor localization technical feasibility has been verified, and its market demands is huge. The channel state information (CSI) extracted from Wi-Fi is physical layer information which is more fine-grained than the received signal strength indication (RSSI). This paper proposes a CSI correction localization algorithm using DenseNet, which is termed CorFi. This method first uses isolation forest to eliminate abnormal CSI, and then constructs a CSI amplitude fingerprint containing time, frequency and antenna pair information. In an offline stage, the densely connected convolutional networks (DenseNet) are trained to establish correspondence between CSI and spatial position, and generalized extended interpolation is applied to construct the interpolated fingerprint database. In an online stage, DenseNet is used for position estimation, and the interpolated fingerprint database and K-nearest neighbor (KNN) are combined to correct the position of the prediction results with low maximum probability. In an indoor corridor environment, the average localization error is 0.536m.

In this paper, we propose a novel user selection scheme based on jointly combining channel gain (CG) and signal to interference plus noise ratio (SINR) to improve the sum-rate as well as to reduce the computation complexity of multi-user massive multi-input multi-output (MU-massive MIMO) downlink transmission through a block diagonalization (BD) precoding technique. By jointly considering CG and SINR based user sets, sum-rate performance improvement can be achieved by selecting higher gain users with better SINR conditions as well as by eliminating the users who cause low sum-rate in the system. Through this approach, the number of possible outcomes for the user selection scheme can be reduced by counting the common users for every pair of user combinations in the selection process since the common users of CG-based and SINR-based sets possess both higher channel gains and better SINR conditions. The common users set offers not only sum-rate performance improvements but also computation complexity reduction in the proposed scheme. It is shown by means of computer simulation experiments that the proposed scheme can increase the sum-rate with lower computation complexity for various numbers of users as compared to conventional schemes requiring the same or less computational complexity.

This paper presents a simulation study on the printing behavior of three different EUV resist systems. Stochastic models for negative metal-based resist and conventional chemically amplified resist (CAR) were calibrated and then validated. As for negative-tone development (NTD) CAR, we commenced from a positive-tone development (PTD) CAR calibrated (material) and NTD development models, since state-of-the-art measurements are not available. A conceptual study between PTD CAR and NTD CAR shows that the stochastic inhibitor fluctuation differs for PTD CAR: the inhibitor level exhibits small fluctuation (Mack development). For NTD CAR, the inhibitor fluctuation depends on the NTD type, which is defined by categorizing the difference between the NTD and PTD development thresholds. Respective NTD types have different inhibitor concentration level. Moreover, contact hole printing between negative metal-based and NTD CAR was compared to clarify the stochastic process window (PW) for tone reversed mask. For latter comparison, the aerial image (AI) and secondary electron effect are comparable. Finally, the local CD uniformity (LCDU) for the same 20 nm size, 40 nm pitch contact hole was compared among the three different resists. Dose-dependent behavior of LCDU and stochastic PW for NTD were different for the PTD CAR and metal-based resist. For NTD CAR, small inhibitor level and large inhibitor fluctuation around the development threshold were observed, causing LCDU increase, which is specific to the inverse Mack development resist.

In infrastructure-as-a-service platforms, cloud users can adjust their database (DB) service scale to dynamic workloads by changing the number of virtual machines running a DB management system (DBMS), called DBMS instances. Replicating a DBMS instance is a non-trivial task since DBMS replication is time-consuming due to the trend that cloud vendors offer high-spec DBMS instances. This paper presents BalenaDB, which performs urgent DBMS replication for handling sudden workload increases. Unlike convectional replication schemes that implicitly assume DBMS replicas are generated on remote machines, BalenaDB generates a warmed-up DBMS replica on an instance running on the local machine where the master DBMS instance runs, by leveraging the master DBMS resources. We prototyped BalenaDB on MySQL 5.6.21, Linux 3.17.2, and Xen 4.4.1. The experimental results show that the time for generating the warmed-up DBMS replica instance on BalenaDB is up to 30× shorter than an existing DBMS instance replication scheme, achieving significantly efficient memory utilization.

Local discriminative regions play important roles in fine-grained image analysis tasks. How to locate local discriminative regions with only category label and learn discriminative representation from these regions have been hot spots. In our work, we propose Searching Discriminative Regions (SDR) and Learning Discriminative Regions (LDR) method to search and learn local discriminative regions in images. The SDR method adopts attention mechanism to iteratively search for high-response regions in images, and uses this as a clue to locate local discriminative regions. Moreover, the LDR method is proposed to learn compact within category and sparse between categories representation from the raw image and local images. Experimental results show that our proposed approach achieves excellent performance in both fine-grained image retrieval and classification tasks, which demonstrates its effectiveness.

Most conventional multi-source domain adaptation techniques for recurrent neural network language models (RNNLMs) are domain-centric. In these approaches, each domain is considered independently and this makes it difficult to apply the models to completely unseen target domains that are unobservable during training. Instead, our study exploits domain attributes, which represent common knowledge among such different domains as dialects, types of wordings, styles, and topics, to achieve domain generalization that can robustly represent unseen target domains by combining the domain attributes. To achieve attribute-based domain generalization system in language modeling, we introduce domain attribute-based experts to a multi-stream RNNLM called recurrent adaptive mixture model (RADMM) instead of domain-based experts. In the proposed system, a long short-term memory is independently trained on each domain attribute as an expert model. Then by integrating the outputs from all the experts in response to the context-dependent weight of the domain attributes of the current input, we predict the subsequent words in the unseen target domain and exploit the specific knowledge of each domain attribute. To demonstrate the effectiveness of our proposed domain attributes-centric language model, we experimentally compared the proposed model with conventional domain-centric language model by using texts taken from multiple domains including different writing styles, topics, dialects, and types of wordings. The experimental results demonstrated that lower perplexity can be achieved using domain attributes.

Unsupervised Feature selection is an important dimensionality reduction technique to cope with high-dimensional data. It does not require prior label information, and has recently attracted much attention. However, it cannot fully utilize the discriminative information of samples, which may affect the feature selection performance. To tackle this problem, in this letter, we propose a novel discriminative virtual label regression method (DVLR) for unsupervised feature selection. In DVLR, we develop a virtual label regression function to guide the subspace learning based feature selection, which can select more discriminative features. Moreover, a linear discriminant analysis (LDA) term is used to make the model be more discriminative. To further make the model be more robust and select more representative features, we impose the ℓ2,1-norm on the regression and feature selection terms. Finally, extensive experiments are carried out on several public datasets, and the results demonstrate that our proposed DVLR achieves better performance than several state-of-the-art unsupervised feature selection methods.

Deep neural networks (DNNs) have achieved excellent performance on several tasks and have been widely applied in both academia and industry. However, DNNs are vulnerable to adversarial machine learning attacks in which noise is added to the input to change the networks' output. Consequently, DNN-based mission-critical applications such as those used in self-driving vehicles have reduced reliability and could cause severe accidents and damage. Moreover, adversarial examples could be used to poison DNN training data, resulting in corruptions of trained models. Besides the need for detecting adversarial examples, correcting them is important for restoring data and system functionality to normal. We have developed methods for detecting and correcting adversarial images that use multiple image processing operations with multiple parameter values. For detection, we devised a statistical-based method that outperforms the feature squeezing method. For correction, we devised a method that uses for the first time two levels of correction. The first level is label correction, with the focus on restoring the adversarial images' original predicted labels (for use in the current task). The second level is image correction, with the focus on both the correctness and quality of the corrected images (for use in the current and other tasks). Our experiments demonstrated that the correction method could correct nearly 90% of the adversarial images created by classical adversarial attacks and affected only about 2% of the normal images.