New articles on Electrical Engineering and Systems Science


[1] 2208.07371

Fast localization and single-pixel imaging of the moving object using time-division multiplexing

When imaging moving objects, single-pixel imaging produces motion blur. This paper proposes a new single-pixel imaging method, which can achieve anti-motion blur imaging of a fast-moving object. The geometric moment patterns and Hadamard patterns are used to alternately encode the position information and the image information of the object with time-division multiplexing. In the reconstruction process, the object position information is extracted independently and combining motion-compensation reconstruction algorithm to decouple the object motion from image information. As a result, the anti-motion blur image and the high frame rate object positions are obtained. Experimental results show that for a moving object with an angular velocity of up to 0.5rad/s relative to the imaging system, the proposed method achieves a localization frequency of 5.55kHz, and gradually reconstructs a clear image of the fast-moving object with a pseudo resolution of 512x512. The method has application prospects in single-pixel imaging of the fast-moving object.


[2] 2208.07414

Progress towards machine learning methodologies for laser-induced breakdown spectroscopy with an emphasis on soil analysis

Optical emission spectroscopy of laser-produced plasmas, commonly known as laser-induced breakdown spectroscopy (LIBS), is an emerging analytical tool for rapid soil analysis. However, specific challenges with LIBS exist, such as matrix effects and quantification issues, that require further study in the application of LIBS, particularly for analysis of heterogeneous samples such as soils. Advancements in the applications of Machine Learning (ML) methods can address some of these issues, advancing the potential for LIBS in soil analysis. This article aims to review the progress of LIBS application combined with ML methods, focusing on methodological approaches used in reducing matrix effect, feature selection, quantification analysis, soil classification, and self-absorption. The performance of various adopted ML approaches is discussed, including their shortcomings and advantages, to provide researchers with a clear picture of the current status of ML applications in LIBS for improving its analytical capability. The challenges and prospects of LIBS development in soil analysis are proposed, offering a path toward future research. This review article emphasize ML tools for LIBS soil analysis that are broadly relevant for other LIBS applications.


[3] 2208.07417

An Efficient Multi-Scale Fusion Network for 3D Organ at Risk (OAR) Segmentation

Accurate segmentation of organs-at-risks (OARs) is a precursor for optimizing radiation therapy planning. Existing deep learning-based multi-scale fusion architectures have demonstrated a tremendous capacity for 2D medical image segmentation. The key to their success is aggregating global context and maintaining high resolution representations. However, when translated into 3D segmentation problems, existing multi-scale fusion architectures might underperform due to their heavy computation overhead and substantial data diet. To address this issue, we propose a new OAR segmentation framework, called OARFocalFuseNet, which fuses multi-scale features and employs focal modulation for capturing global-local context across multiple scales. Each resolution stream is enriched with features from different resolution scales, and multi-scale information is aggregated to model diverse contextual ranges. As a result, feature representations are further boosted. The comprehensive comparisons in our experimental setup with OAR segmentation as well as multi-organ segmentation show that our proposed OARFocalFuseNet outperforms the recent state-of-the-art methods on publicly available OpenKBP datasets and Synapse multi-organ segmentation. Both of the proposed methods (3D-MSF and OARFocalFuseNet) showed promising performance in terms of standard evaluation metrics. Our best performing method (OARFocalFuseNet) obtained a dice coefficient of 0.7995 and hausdorff distance of 5.1435 on OpenKBP datasets and dice coefficient of 0.8137 on Synapse multi-organ segmentation dataset.


[4] 2208.07424

Deep Reinforcement Learning for RIS-Assisted FD Systems: Single or Distributed RIS?

This paper investigates reconfigurable intelligent surface (RIS)-assisted full-duplex multiple-input single-output wireless system, where the beamforming and RIS phase shifts are optimized to maximize the sum-rate for both single and distributed RIS deployment schemes. The preference of using the single or distributed RIS deployment scheme is investigated through three practical scenarios based on the links' quality. The closed-form solution is derived to optimize the beamforming vectors and a novel deep reinforcement learning (DRL) algorithm is proposed to optimize the RIS phase shifts. Simulation results illustrate that the choice of the deployment scheme depends on the scenario and the links' quality. It is further shown that the proposed algorithm significantly improves the sum-rate compared to the non-optimized scenario in both single and distributed RIS deployment schemes. Besides, the proposed beamforming derivation achieves a remarkable improvement compared to the approximated derivation in previous works. Finally, the complexity analysis confirms that the proposed DRL algorithm reduces the computation complexity compared to the DRL algorithm in the literature.


[5] 2208.07437

Retrospective Cost Parameter Estimation with Application to Space Weather Modeling

This chapter reviews standard parameter-estimation techniques and presents a novel gradient-, ensemble-, adjoint-free data-driven parameter estimation technique in the DDDAS framework. This technique, called retrospective cost parameter estimation (RCPE), is motivated by large-scale complex estimation models characterized by high-dimensional nonlinear dynamics, nonlinear parameterizations, and representational models. RCPE is illustrated by estimating unknown parameters in three examples. In the first example, salient features of RCPE are investigated by considering parameter estimation problem in a low-order nonlinear system. In the second example, RCPE is used to estimate the convective coefficient and the viscosity in the generalized Burgers equation by using a scalar measurement. In the final example, RCPE is used to estimate thermal conductivity coefficients that relate temporal temperature variation with the vertical gradient of the temperature in the atmosphere.


[6] 2208.07446

C3-DINO: Joint Contrastive and Non-contrastive Self-Supervised Learning for Speaker Verification

Self-supervised learning (SSL) has drawn an increased attention in the field of speech processing. Recent studies have demonstrated that contrastive learning is able to learn discriminative speaker embeddings in a self-supervised manner. However, base contrastive self-supervised learning (CSSL) assumes that the pairs generated from a view of anchor instance and any view of other instances are all negative, which introduces many false negative pairs in constructing the loss function. The problem is referred as $class$-$collision$, which remains as one major issue that impedes the CSSL based speaker verification (SV) systems from achieving better performances. In the meanwhile, studies reveal that negative sample free SSL frameworks perform well in learning speaker or image representations. In this study, we investigate SSL techniques that lead to an improved SV performance. We first analyse the impact of false negative pairs in the CSSL systems. Then, a multi-stage Class-Collision Correction (C3) method is proposed, which leads to the state-of-the-art CSSL based speaker embedding system. On the basis of the pretrained CSSL model, we further propose to employ a negative sample free SSL objective (i.e., DINO) to fine-tune the speaker embedding network. The resulting speaker embedding system (C3-DINO) achieves 2.5% EER with a simple Cosine Distance Scoring method on Voxceleb1 test set, which outperforms the previous SOTA SSL system (4.86%) by a significant +45% relative improvement. With speaker clustering and pseudo labeling on Voxceleb2 training set, a LDA/CDS back-end applying on the C3-DINO speaker embeddings is able to further push the EER to 2.2%. Comprehensive experimental investigations of the Voxceleb benchmarks and our internal dataset demonstrate the effectiveness of our proposed methods, and the performance gap between the SSL SV and the supervised counterpart narrows further.


[7] 2208.07485

Core-shell enhanced single particle model for lithium iron phosphate batteries: model formulation and analysis of numerical solutions

In this paper, a core-shell enhanced single particle model for iron-phosphate battery cells is formulated, implemented, and verified. Starting from the description of the positive and negative electrodes charge and mass transport dynamics, the positive electrode intercalation and deintercalation phenomena and associated phase transitions are described with the core-shell modeling paradigm. Assuming two phases are formed in the positive electrode, one rich and one poor in lithium, a core-shrinking problem is formulated and the phase transition is modeled through a shell phase that covers the core one. A careful discretization of the coupled partial differential equations is proposed and used to convert the model into a system of ordinary differential equations. To ensure robust and accurate numerical solutions of the governing equations, a sensitivity analysis of numerical solutions is performed and the best setting, in terms of solver tolerances, solid phase concentration discretization points, and input current sampling time, is determined in a newly developed probabilistic framework. Finally, unknown model parameters are identified at different C-rate scenarios and the model is verified against experimental data.


[8] 2208.07510

EM and SAGE algorithms for DOA Estimation in the Presence of Unknown Uniform Noise

The expectation-maximization (EM) and space-alternating generalized EM (SAGE) algorithms have been applied to direction of arrival (DOA) estimation in known noise. In this work, the two algorithms are proposed for DOA estimation in unknown uniform noise. Both the deterministic and stochastic signal models are considered. Moreover, a modified EM (MEM) algorithm applicable to the noise assumption is also proposed. These proposed algorithms are improved to ensure the stability when the powers of sources are unequal. After being improved, numerical results illustrate that the EM algorithm has similar convergence with the MEM algorithm and the SAGE algorithm outperforms the EM and MEM algorithms for the deterministic signal model. Furthermore, numerical results show that processing the same samples from the stochastic signal model, the SAGE algorithm for the deterministic signal model requires the fewest iterations.


[9] 2208.07513

Evaluate Quantum Combinatorial Optimization for Distribution Network Reconfiguration

This paper aims to implement and evaluate the performance of quantum computing on solving combinatorial optimization problems arising from the operations of the power grid. To this end, we construct a novel mixed integer conic programming formulation for the reconfiguration of radial distribution network in response to faults in distribution lines. Comparing to existing bus injection model in the literature, our formulation based the branch flows model is theoretically equivalent without needing non-explainable variables, thus being more numerically stable. The network reconfiguration model is then used as a benchmark to evaluate the performance of quantum computing algorithms in real quantum computers. It shows that while current quantum computing algorithms with fast execution time in quantum computers can be a promising solution candidate, its heuristic nature stem from its theoretical foundation should be considered carefully when applying into power grid optimization problems.


[10] 2208.07525

Design and Analysis of Hardware-limited Non-uniform Task-based Quantizers

Hardware-limited task-based quantization is a new design paradigm for data acquisition systems equipped with serial scalar analog-to-digital converters using a small number of bits. By taking into account the underlying system task, task-based quantizers can efficiently recover the desired parameters from the low-bit quantized observation. Current design and analysis frameworks for hardware-limited task-based quantization are only applicable to inputs with bounded support and uniform quantizers with non-subtractive dithering. Here, we propose a new framework based on generalized Bussgang decomposition that enables the design and analysis of hardware-limited task-based quantizers that are equipped with non-uniform scalar quantizers or that have inputs with unbounded support. We first consider the scenario in which the task is linear. Under this scenario, we derive new pre-quantization and post-quantization linear mappings for task-based quantizers with mean squared error (MSE) that closely matches the theoretical MSE. Next, we extend the proposed analysis framework to quadratic tasks. We demonstrate that our derived analytical expression for the MSE accurately predicts the performance of task-based quantizers with quadratic tasks.


[11] 2208.07528

Integrating Satellites and Mobile Edge Computing for 6G Wide-Area Edge Intelligence: Minimal Structures and Systematic Thinking

The sixth-generation (6G) network will shift its focus to supporting everything including various machine-type devices (MTDs) in an everyone-centric manner. To ubiquitously cover the MTDs working in rural and disastrous areas, satellite communications become indispensable, while mobile edge computing (MEC) also plays an increasingly crucial role. Their sophisticated integration enables wide-area edge intelligence which promises to facilitate globally-distributed customized services. In this article, we present typical use cases of integrated satellite-MEC networks and discuss the main challenges therein. Inspired by the protein structure and the systematic engineering methodology, we propose three minimal integrating structures, based on which a complex integrated satellite-MEC network can be treated as their extension and combination. We discuss the unique characteristics and key problems of each minimal structure. Accordingly, we establish an on-demand network orchestration framework to enrich the hierarchy of network management, which further leads to a process-oriented network optimization method. On that basis, a case study is utilized to showcase the benefits of on-demand network orchestration and process-oriented network optimization. Finally, we outline potential research issues to envision a more intelligent, more secure, and greener integrated network.


[12] 2208.07550

Hybrid UAV-enabled Secure Offloading via Deep Reinforcement Learning

Unmanned aerial vehicles (UAVs) have been actively studied as moving cloudlets to provide application offloading opportunities and to enhance the security level of user equipments (UEs). In this correspondence, we propose a hybrid UAV-aided secure offloading system in which a UAV serves as a helper by switching the mode between jamming and relaying to maximize the secrecy sum-rate of UEs. This work aims to optimize (i) the trajectory of the helper UAV, (ii) the mode selection strategy and (iii) the UEs' offloading decisions under the constraints of offloading accomplishment and the UAV's operational limitations. The solution is provided via a deep deterministic policy gradient (DDPG)-based method, whose superior performance is verified via a numerical simulation and compared to those of traditional approaches.


[13] 2208.07552

Coil2Coil: Self-supervised MR image denoising using phased-array coil images

Denoising of magnetic resonance images is beneficial in improving the quality of low signal-to-noise ratio images. Recently, denoising using deep neural networks has demonstrated promising results. Most of these networks, however, utilize supervised learning, which requires large training images of noise-corrupted and clean image pairs. Obtaining training images, particularly clean images, is expensive and time-consuming. Hence, methods such as Noise2Noise (N2N) that require only pairs of noise-corrupted images have been developed to reduce the burden of obtaining training datasets. In this study, we propose a new self-supervised denoising method, Coil2Coil (C2C), that does not require the acquisition of clean images or paired noise-corrupted images for training. Instead, the method utilizes multichannel data from phased-array coils to generate training images. First, it divides and combines multichannel coil images into two images, one for input and the other for label. Then, they are processed to impose noise independence and sensitivity normalization such that they can be used for the training images of N2N. For inference, the method inputs a coil-combined image (e.g., DICOM image), enabling a wide application of the method. When evaluated using synthetic noise-added images, C2C shows the best performance against several self-supervised methods, reporting comparable outcomes to supervised methods. When testing the DICOM images, C2C successfully denoised real noise without showing structure-dependent residuals in the error maps. Because of the significant advantage of not requiring additional scans for clean or paired images, the method can be easily utilized for various clinical applications.


[14] 2208.07566

Multi-dimensional topological loss for cortical plate segmentation in fetal brain MRI

The fetal cortical plate (CP) undergoes drastic morphological changes during the in utero development. Therefore, CP growth and folding patterns are key indicator in the assessment of the brain development and maturation. Magnetic resonance imaging (MRI) offers specific insights for the analysis of quantitative imaging biomarkers. Nonetheless, accurate and, more importantly, topologically correct MR image segmentation remains the key baseline to such analysis. In this study, we propose a deep learning segmentation framework for automatic and morphologically consistent segmentation of the CP in fetal brain MRI. Our contribution is two fold. First, we generalized a multi-dimensional topological loss function in order to enhance the topological accuracy. Second, we introduced hole ratio, a new topology-based validation measure that quantifies the size of the topological defects taking into account the size of the structure of interest. Using two publicly available datasets, we quantitatively evaluated our proposed method based on three complementary metrics which are overlap-, distance- and topology-based on 27 fetal brains. Our results evidence that our topology-integrative framework outperforms state-of-the-art training loss functions on super-resolution reconstructed clinical MRI, not only in shape correctness but also in the classical evaluation metrics. Furthermore, results on additional 31 out-of-domain SR reconstructions from clinical acquisitions were qualitatively assessed by three experts. The experts' consensus ranked our TopoCP method as the best segmentation in 100\% of the cases with a high inter-expert agreement. Overall, both quantitative and qualitative results, on a wide range of gestational ages and number of cases, support the generalizability and added value of our topology-guided framework for fetal CP segmentation.


[15] 2208.07593

The Low Emission Oil&Gas Open (LEOGO) Reference Platform of an Off-Grid Energy System for Renewable Integration Studies

This article introduces and describes the integrated energy system of a the Low Emission Oil&Gas Open (LEOGO) reference platform. It is a hypothetical case meant to represent a typical oil&gas installation in the North Sea. The aim of this detailed specification is to serve as an open reference case where all information about it can be publicly shared, facilitating benchmarking and collaboration. The relevance of this reference case of an off-grid energy system is not limited to the oil&gas industry, since it can also been seen as a special kind of electrical micro grid. The remote offshore location makes it especially relevant for studying offshore wind power and ocean energy sources like wave power. The specification has an emphasis on the energy system and electrical configuration, but includes also a basic description of oil field and processing system. The intention is that it will serve as a basis for energy system studies and relating power system stability analyses regarding the integration of renewable energy sources. This allows for comparisons of a base case with different design modifications, new operational planning methods, power management strategies and control concepts. Examples of possible modifications are the replacement of gas turbines by wind turbines, addition of energy storage systems, a more variable operation of loads, etc. The last part of the article demonstrates the behaviour of the reference platform implemented in two software tools. One for operational planning and one for dynamic power system analyses.


[16] 2208.07602

Two-Step mmWave Positioning Scheme with RIS-Part I: Angle Estimation and Analysis

In this series of work, we propose a comprehensive two-step three-dimensional (3D) positioning scheme in a millimeter wave (mmWave) system, where the reconfigurable intelligent surface (RIS) is leveraged to enhance the positioning performance of mobile users (MUs). Specifically, the first step is the estimation error modeling and analysis based on the two-dimensional discrete Fourier transform (2D-DFT) angle estimation technique, while the second step is the corresponding positioning algorithm design and bias analysis. The first step is introduced in this paper, and the second step is investigated in Part II of this series work. Based on 2D-DFT angle estimation, the angle estimation error is modeled and analyzed by deriving its probability density functions (PDF). More specifically, we first derive the PDF by using the geometric relationship between the angles of arrival (AOAs) and their triangle functions. Then, we simplify the intricate expression of the PDF of the AOA estimation error by employing the first-order linear approximation of triangle functions. Finally, we derive a complex expression for the variance based on the derived PDF. Distinctively, for the azimuth estimation error, the variance is separately integrated according to the different non-zero intervals of the PDF, which will be used in the second part of this series work for the analysis of the position estimation error. Extensive simulation results are also presented to verify the accuracy of the derived results.


[17] 2208.07603

Online Learning Based NLOS Ranging Error Mitigation in 5G Positioning

The fifth-generation (5G) wireless communication is useful for positioning due to its large bandwidth and low cost. However, the presence of obstacles that block the line-of-sight (LOS) path between devices would affect localization accuracy severely. In this paper, we propose an online learning approach to mitigate ranging error directly in non-line-of-sight (NLOS) channels. The distribution of NLOS ranging error is learned from received raw signals, where a network with neural processes regressor (NPR) is utilized to learn the environment and range-related information precisely. The network can be implemented for online learning free from retraining the network, which is computationally efficient. Simulation results show that the proposed approach outperforms conventional techniques in terms of NLOS ranging error mitigation.


[18] 2208.07606

Two-Step mmWave Positioning Scheme with RIS-Part II: Position Estimation and Error Analysis

In this series of work, we propose a comprehensive two-step three-dimensional (3D) positioning scheme in a millimeter wave (mmWave) system, where the reconfigurable intelligent surface (RIS) is leveraged to enhance the positioning performance of mobile users (MUs). Specifically, the first step is the estimation error modeling and analysis, while the second step is the corresponding positioning algorithm design and bias analysis. The first step is introduced in Part I of this series of work, and the second step is investigated in this paper. Our aim in this series of work is to obtain the closed-form solution of the MU's position through a two-stage weight least square (TSWLS) algorithm. In the first stage, we construct the pseudolinear equations based on the angle of arrival (AOA) and the time difference of arrival (TDOA) estimation at the RISs, then we obtain a preliminary estimation by solving these equations using the weight least square (WLS) method. Based on the preliminary estimation in the first stage, a new set of pseudolinear equations are obtained, and a finer estimation is obtained by solving the equations using the WLS method in the second stage. By combining the estimation of both stages, the final estimation of the MU's position is obtained. Further, we study the theoretical bias of the proposed algorithm by considering the estimation error in both stages. Simulation results demonstrate the superiority of the proposed positioning scheme.


[19] 2208.07607

Vacuum Circuit Breaker Closing Time Key Moments Detection via Vibration Monitoring: A Run-to-Failure Study

Circuit breakers (CBs) play an important role in modern society because they make the power transmission and distribution systems reliable and resilient. Therefore, it is important to maintain their reliability and to monitor their operation. A key to ensure a reliable operation of CBs is to monitor their condition. In this work, we performed an accelerated life testing for mechanical failures of a vacuum circuit breaker (VCB) by performing close-open operations continuously until failure. We recorded data for each operation and made the collected run-to-failure dataset publicly available. In our experiments, the VCB operated more than 26000 close-open operations without current load with the time span of five months. The run-to-failure long-term monitoring enables us to monitor the evolution of the VCB condition and the degradation over time. To monitor CB condition, closing time is one of the indicators, which is usually measured when the CB is taken out of operation and is completely disconnected from the network. We propose an algorithm that enables to infer the same information on the closing time from a non-intrusive sensor. By utilizing the short-time energy (STE) of the vibration signal, it is possible to identify the key moments when specific events happen including the time when the latch starts to move, and the closing time. The effectiveness of the proposed algorithm is evaluated on the VCB dataset and is also compared to the binary segmentation (BS) change point detection algorithm. This research highlights the potential for continuous online condition monitoring, which is the basis for applying future predictive maintenance strategies.


[20] 2208.07639

RAWtoBit: A Fully End-to-end Camera ISP Network

Image compression is an essential and last processing unit in the camera image signal processing (ISP) pipeline. While many studies have been made to replace the conventional ISP pipeline with a single end-to-end optimized deep learning model, image compression is barely considered as a part of the model. In this paper, we investigate the designing of a fully end-to-end optimized camera ISP incorporating image compression. To this end, we propose RAWtoBit network (RBN) that can effectively perform both tasks simultaneously. RBN is further improved with a novel knowledge distillation scheme by introducing two teacher networks specialized in each task. Extensive experiments demonstrate that our proposed method significantly outperforms alternative approaches in terms of rate-distortion trade-off.


[21] 2208.07647

Grapes disease detection using transfer learning

Early and precise diagnosis of diseases in plants can help to develop an early treatment technique. Plant diseases degrade both the quantity and quality of crops, thus posing a threat to food security and resulting in huge economic losses. Traditionally identification is performed manually, which is inaccurate, time-consuming, and expensive. This paper presents a simple and efficient model to detect grapes leaf diseases using transfer learning. A pre-trained deep convolutional neural network is used as a feature extractor and random forest as a classifier. The performance of the model is interpreted in terms of accuracy, precision, recall, and f1 score. Total 1003 images of four different classes are used and 91.66% accuracy is obtained.


[22] 2208.07655

A Hybrid Deep Feature-Based Deformable Image Registration Method for Pathological Images

Pathologists need to combine information from differently stained pathological slices to obtain accurate diagnostic results. Deformable image registration is a necessary technique for fusing multi-modal pathological slices. This paper proposes a hybrid deep feature-based deformable image registration framework for stained pathological samples. We first extract dense feature points and perform points matching by two deep learning feature networks. Then, to further reduce false matches, an outlier detection method combining the isolation forest statistical model and the local affine correction model is proposed. Finally, the interpolation method generates the DVF for pathology image registration based on the above matching points. We evaluate our method on the dataset of the Non-rigid Histology Image Registration (ANHIR) challenge, which is co-organized with the IEEE ISBI 2019 conference. Our technique outperforms the traditional approaches by 17% with the Average-Average registration target error (rTRE) reaching 0.0034. The proposed method achieved state-of-the-art performance and ranking it 1 in evaluating the test dataset. The proposed hybrid deep feature-based registration method can potentially become a reliable method for pathology image registration.


[23] 2208.07657

Uconv-Conformer: High Reduction of Input Sequence Length for End-to-End Speech Recognition

Optimization of modern ASR architectures is among the highest priority tasks since it saves many computational resources for model training and inference. The work proposes a new Uconv-Conformer architecture based on the standard Conformer model that consistently reduces the input sequence length by 16 times, which results in speeding up the work of the intermediate layers. To solve the convergence problem with such a significant reduction of the time dimension, we use upsampling blocks similar to the U-Net architecture to ensure the correct CTC loss calculation and stabilize network training. The Uconv-Conformer architecture appears to be not only faster in terms of training and inference but also shows better WER compared to the baseline Conformer. Our best Uconv-Conformer model showed 40.3% epoch training time reduction, 47.8%, and 23.5% inference acceleration on the CPU and GPU, respectively. Relative WER on Librispeech test_clean and test_other decreased by 7.3% and 9.2%.


[24] 2208.07673

A non-invasive fault location method for modular multilevel converters under light load conditions

This paper proposes a non-invasive fault location method for modular multilevel converters (MMC) considering light load conditions. The prior-art fault location methods of the MMC are often developed and verified under full load conditions. However, it is revealed that the faulty arm current will be suppressed to be unipolar when the open-circuit fault happens on the submodule switch under light load. This leads to the capacitor voltage of the healthy and faulty submodules rising or falling with the same variations, increasing the difficulty of fault location. The proposed approach of injecting the second-order circulating current will rebuild the bipolar arm current of the MMC and enlarge the capacitor voltage deviations between the healthy and faulty SMs. As a result, the fault location time is significantly shortened. The simulations are carried out to validate the effectiveness of the proposed approach, showing that the fault location time is reduced to 1/6 compared with the condition without second-order circulating current injection.


[25] 2208.07739

A Latent Feature Analysis-based Approach for Spatio-Temporal Traffic Data Recovery

Missing data is an inevitable and common problem in data-driven intelligent transportation systems (ITS). In the past decade, scholars have done many research on the recovery of missing traffic data, however how to make full use of spatio-temporal traffic patterns to improve the recovery performance is still an open problem. Aiming at the spatio-temporal characteristics of traffic speed data, this paper regards the recovery of missing data as a matrix completion problem, and proposes a spatio-temporal traffic data completion method based on hidden feature analysis, which discovers spatio-temporal patterns and underlying structures from incomplete data to complete the recovery task. Therefore, we introduce spatial and temporal correlation to capture the main underlying features of each dimension. Finally, these latent features are applied to recovery traffic data through latent feature analysis. The experimental and evaluation results show that the evaluation criterion value of the model is small, which indicates that the model has better performance. The results show that the model can accurately estimate the continuous missing data.


[26] 2208.07787

Delaunay-Triangulation-Based Learning with Hessian Total-Variation Regularization

Regression is one of the core problems tackled in supervised learning. Rectified linear unit (ReLU) neural networks generate continuous and piecewise-linear (CPWL) mappings and are the state-of-the-art approach for solving regression problems. In this paper, we propose an alternative method that leverages the expressivity of CPWL functions. In contrast to deep neural networks, our CPWL parameterization guarantees stability and is interpretable. Our approach relies on the partitioning of the domain of the CPWL function by a Delaunay triangulation. The function values at the vertices of the triangulation are our learnable parameters and identify the CPWL function uniquely. Formulating the learning scheme as a variational problem, we use the Hessian total variation (HTV) as regularizer to favor CPWL functions with few affine pieces. In this way, we control the complexity of our model through a single hyperparameter. By developing a computational framework to compute the HTV of any CPWL function parameterized by a triangulation, we discretize the learning problem as the generalized least absolute shrinkage and selection operator (LASSO). Our experiments validate the usage of our method in low-dimensional scenarios.


[27] 2208.07829

Diagnosis of COVID-19 disease using CT scan images and pre-trained models

Diagnosis of COVID-19 is necessary to prevent and control the disease. Deep learning methods have been considered a fast and accurate method. In this paper, by the parallel combination of three well-known pre-trained networks, we attempted to distinguish coronavirus-infected samples from healthy samples. The negative log-likelihood loss function has been used for model training. CT scan images in the SARS-CoV-2 dataset were used for diagnosis. The SARS-CoV-2 dataset contains 2482 images of lung CT scans, of which 1252 images belong to COVID-19-infected samples. The proposed model was close to 97% accurate.


[28] 2208.07830

Signal detection with dynamic programming

This paper studies the classical problem of detecting the locations of signal occurrences in a one-dimensional noisy measurement. Assuming the signal occurrences do not overlap, we formulate the detection task as a constrained likelihood optimization problem, and design a computationally efficient dynamic program that attains its optimal solution. Our proposed framework is scalable, simple to implement, and robust to model uncertainties. We show by extensive numerical experiments that our algorithm accurately estimates the locations in dense and noisy environments, and outperforms alternative methods.


[29] 2208.07855

Deep Learning for Size and Microscope Feature Extraction and Classification in Oral Cancer: Enhanced Convolution Neural Network

Background and Aim: Over-fitting issue has been the reason behind deep learning technology not being successfully implemented in oral cancer images classification. The aims of this research were reducing overfitting for accurately producing the required dimension reduction feature map through Deep Learning algorithm using Convolutional Neural Network. Methodology: The proposed system consists of Enhanced Convolutional Neural Network that uses an autoencoder technique to increase the efficiency of the feature extraction process and compresses information. In this technique, unpooling and deconvolution is done to generate the input data to minimize the difference between input and output data. Moreover, it extracts characteristic features from the input data set to regenerate input data from those features by learning a network to reduce overfitting. Results: Different accuracy and processing time value is achieved while using different sample image group of Confocal Laser Endomicroscopy (CLE) images. The results showed that the proposed solution is better than the current system. Moreover, the proposed system has improved the classification accuracy by 5~ 5.5% on average and reduced the average processing time by 20 ~ 30 milliseconds. Conclusion: The proposed system focuses on the accurate classification of oral cancer cells of different anatomical locations from the CLE images. Finally, this study enhances the accuracy and processing time using the autoencoder method that solves the overfitting problem.


[30] 2208.07422

Deep Unsupervised Domain Adaptation: A Review of Recent Advances and Perspectives

Deep learning has become the method of choice to tackle real-world problems in different domains, partly because of its ability to learn from data and achieve impressive performance on a wide range of applications. However, its success usually relies on two assumptions: (i) vast troves of labeled datasets are required for accurate model fitting, and (ii) training and testing data are independent and identically distributed. Its performance on unseen target domains, thus, is not guaranteed, especially when encountering out-of-distribution data at the adaptation stage. The performance drop on data in a target domain is a critical problem in deploying deep neural networks that are successfully trained on data in a source domain. Unsupervised domain adaptation (UDA) is proposed to counter this, by leveraging both labeled source domain data and unlabeled target domain data to carry out various tasks in the target domain. UDA has yielded promising results on natural image processing, video analysis, natural language processing, time-series data analysis, medical image analysis, etc. In this review, as a rapidly evolving topic, we provide a systematic comparison of its methods and applications. In addition, the connection of UDA with its closely related tasks, e.g., domain generalization and out-of-distribution detection, has also been discussed. Furthermore, deficiencies in current methods and possible promising directions are highlighted.


[31] 2208.07448

Self-Supervised Learning for Anomalous Channel Detection in EEG Graphs: Application to Seizure Analysis

Electroencephalogram (EEG) signals are effective tools towards seizure analysis where one of the most important challenges is accurate detection of seizure events and brain regions in which seizure happens or initiates. However, all existing machine learning-based algorithms for seizure analysis require access to the labeled seizure data while acquiring labeled data is very labor intensive, expensive, as well as clinicians dependent given the subjective nature of the visual qualitative interpretation of EEG signals. In this paper, we propose to detect seizure channels and clips in a self-supervised manner where no access to the seizure data is needed. The proposed method considers local structural and contextual information embedded in EEG graphs by employing positive and negative sub-graphs. We train our method through minimizing contrastive and generative losses. The employ of local EEG sub-graphs makes the algorithm an appropriate choice when accessing to the all EEG channels is impossible due to complications such as skull fractures. We conduct an extensive set of experiments on the largest seizure dataset and demonstrate that our proposed framework outperforms the state-of-the-art methods in the EEG-based seizure study. The proposed method is the only study that requires no access to the seizure data in its training phase, yet establishes a new state-of-the-art to the field, and outperforms all related supervised methods.


[32] 2208.07497

Active Bucketized Learning for ACOPF Optimization Proxies

This paper considers optimization proxies for Optimal Power Flow (OPF), i.e., machine-learning models that approximate the input/output relationship of OPF. Recent work has focused on showing that such proxies can be of high fidelity. However, their training requires significant data, each instance necessitating the (offline) solving of an OPF for a sample of the input distribution. To meet the requirements of market-clearing applications, this paper proposes Active Bucketized Sampling (ABS), a novel active learning framework that aims at training the best possible OPF proxy within a time limit. ABS partitions the input distribution into buckets and uses an acquisition function to determine where to sample next. It relies on an adaptive learning rate that increases and decreases over time. Experimental results demonstrate the benefits of ABS.


[33] 2208.07536

Dependency Tasks Offloading and Communication Resource Allocation in Collaborative UAVs Networks: A Meta-Heuristic Approach

In recent years, unmanned aerial vehicles (UAVs) assisted mobile edge computing systems have been exploited by researchers as a promising solution for providing computation services to mobile users outside of terrestrial infrastructure coverage. However, it remains challenging for the standalone MEC-enabled UAVs in order to meet the computation requirement of numerous mobile users due to the limited computation capacity of their onboard servers and battery lives. Therefore, we propose a collaborative scheme among UAVs so that UAVs can share the workload with idle UAVs. Moreover, current task offloading strategies frequently overlook task topology, which may result in poor performance or even system failure. To address the problem, we consider offloading tasks consisting of a set of sub-tasks, and each sub-task has dependencies on other sub-tasks, which is practical in the real world. Sub-tasks with dependencies need to wait for the resulting signal from preceding sub-tasks before being executed. This mechanism has serious effects on the offloading strategy. Then, we formulate an optimization problem to minimize the average latency experienced by users by jointly controlling the offloading decision for dependent tasks and allocating the communication resources of UAVs. The formulated problem appears to be NP-hard and cannot be solved in polynomial time. Therefore, we divide the problem into two sub-problems: the offloading decision problem and the communication resource allocation problem. Then a meta-heuristic method is proposed to find the sub-optimal solution of the task offloading problem, while the communication resource allocation problem is solved by using convex optimization. Finally, we perform substantial simulation experiments, and the result shows that the proposed offloading technique effectively minimizes the average latency of users, compared with other benchmark schemes.


[34] 2208.07571

MSE-Based Transceiver Designs for RIS-Aided Communications With Hardware Impairments

It is challenging to precisely configure the phase shifts of the reflecting elements at the reconfigurable intelligent surface (RIS) due to inherent hardware impairments (HIs). In this paper, the mean square error (MSE) performance is investigated in an RIS-aided single-user multiple-input multipleoutput (MIMO) communication system with transceiver HIs and RIS phase noise. We aim to jointly optimize the transmit precoder, linear received equalizer, and RIS reflecting matrices to minimize the MSE. To tackle this problem, an iterative algorithm is proposed, wherein the beamforming matrices are alternately optimized. Specifically, for the beamforming optimization subproblem, we derive the closed-form expression of the optimal precoder and equalizer matrices. Then, for the phase shift optimization subproblem, an efficient algorithm based on the majorization-minimization (MM) method is proposed. Simulation results show that the proposed MSE-based RIS-aided transceiver scheme dramatically outperforms the conventional system algorithms that do not consider HIs at both the transceiver and the RIS.


[35] 2208.07583

HVS-Inspired Signal Degradation Network for Just Noticeable Difference Estimation

Significant improvement has been made on just noticeable difference (JND) modelling due to the development of deep neural networks, especially for the recently developed unsupervised-JND generation models. However, they have a major drawback that the generated JND is assessed in the real-world signal domain instead of in the perceptual domain in the human brain. There is an obvious difference when JND is assessed in such two domains since the visual signal in the real world is encoded before it is delivered into the brain with the human visual system (HVS). Hence, we propose an HVS-inspired signal degradation network for JND estimation. To achieve this, we carefully analyze the HVS perceptual process in JND subjective viewing to obtain relevant insights, and then design an HVS-inspired signal degradation (HVS-SD) network to represent the signal degradation in the HVS. On the one hand, the well learnt HVS-SD enables us to assess the JND in the perceptual domain. On the other hand, it provides more accurate prior information for better guiding JND generation. Additionally, considering the requirement that reasonable JND should not lead to visual attention shifting, a visual attention loss is proposed to control JND generation. Experimental results demonstrate that the proposed method achieves the SOTA performance for accurately estimating the redundancy of the HVS. Source code will be available at https://github.com/jianjin008/HVS-SD-JND.


[36] 2208.07635

A New Scheme for Image Compression and Encryption Using ECIES, Henon Map, and AEGAN

Providing security in the transmission of images and other multimedia data has become one of the most important scientific and practical issues. In this paper, a method for compressing and encryption images is proposed, which can safely transmit images in low-bandwidth data transmission channels. At first, using the autoencoding generative adversarial network (AEGAN) model, the images are mapped to a vector in the latent space with low dimensions. In the next step, the obtained vector is encrypted using public key encryption methods. In the proposed method, Henon chaotic map is used for permutation, which makes information transfer more secure. To evaluate the results of the proposed scheme, three criteria SSIM, PSNR, and execution time have been used.


[37] 2208.07642

A two-step approach to Wasserstein distributionally robust chance- and security-constrained dispatch

This paper considers a security constrained dispatch problem involving generation and line contingencies in the presence of the renewable generation. The uncertainty due to renewables is modeled using joint chance-constraint and the mismatch caused by contingencies and renewables are handled using reserves. We consider a distributionally robust approach to solve the chance-constrained program. We assume that samples of the uncertainty are available. Using them, we construct a set of distributions, termed ambiguity set, containing all distributions that are close to the empirical distribution under the Wasserstein metric. The chance constraint is imposed for all distributions in the ambiguity set to form the distributionally robust optimization problem. This problem is nonconvex and computationally heavy to solve exactly. We adopt a two-step approach to find an approximate solution. In the first step, we construct a polyhedral set in the space of uncertainty that contains enough mass under all distributions in the ambiguity set. This set is constructed by solving several two-dimensional distributionally robust problems. In the second step, we solve a linear robust optimization problem where the uncertain constraint is imposed for all uncertainty values lying in the polyhedral set. We demonstrate the scalability and robustness of our method using numerical experiments.


[38] 2208.07679

How Should We Evaluate Synthesized Environmental Sounds

Although several methods of environmental sound synthesis have been proposed, there has been no discussion on how synthesized environmental sounds should be evaluated. Only either subjective or objective evaluations have been conducted in conventional evaluations, and it is not clear what type of evaluation should be carried out. In this paper, we investigate how to evaluate synthesized environmental sounds. We also propose a subjective evaluation methodology to evaluate whether the synthesized sound appropriately represents the information input to the environmental sound synthesis system. In our experiments, we compare the proposed and conventional evaluation methods and show that the results of subjective evaluations tended to differ from those of objective evaluations. From these results, we conclude that it is necessary to conduct not only objective evaluation but also subjective evaluation.


[39] 2208.07703

A sustainable waste-to-protein system to maximise waste resource utilisation for developing food- and feed-grade protein solutions

A waste-to-protein system that integrates a range of waste-to-protein upgrading technologies has the potential to converge innovations on zero-waste and protein security to ensure a sustainable protein future. We present a global overview of food-safe and feed-safe waste resource potential and technologies to sort and transform such waste streams with compositional quality characteristics into food-grade or feed-grade protein. The identified streams are rich in carbon and nutrients and absent of pathogens and hazardous contaminants, including food waste streams, lignocellulosic waste from agricultural residues and forestry, and contaminant-free waste from the food and drink industry. A wide range of chemical, physical, and biological treatments can be applied to extract nutrients and convert waste-carbon to fermentable sugars or other platform chemicals for subsequent conversion to protein. Our quantitative analyses suggest that the waste-to-protein system has the potential to maximise recovery of various low-value resources and catalyse the transformative solutions toward a sustainable protein future. However, novel protein regulation processes remain expensive and resource intensive in many countries, with protracted timelines for approval. This poses a significant barrier to market expansion, despite accelerated research and development in waste-to-protein technologies and novel protein sources. Thus, the waste-to-protein system is an important initiative to promote metabolic health across the lifespan and tackle the global hunger crisis.


[40] 2208.07735

Rain Removal from Light Field Images with 4D Convolution and Multi-scale Gaussian Process

Existing deraining methods mainly focus on a single input image. With just a single input image, it is extremely difficult to accurately detect rain streaks, remove rain streaks, and restore rain-free images. Compared with a single 2D image, a light field image (LFI) embeds abundant 3D structure and texture information of the target scene by recording the direction and position of each incident ray via a plenoptic camera, which has emerged as a popular device in the computer vision and graphics research communities. In this paper, we propose a novel network, 4D-MGP-SRRNet, for rain streak removal from an LFI. Our method takes as input all sub-views of a rainy LFI. In order to make full use of the LFI, we adopt 4D convolutional layers to build the proposed rain steak removal network to simultaneously process all sub-views of the LFI. In the proposed network, the rain detection model, MGPDNet, with a novel Multi-scale Self-guided Gaussian Process (MSGP) module is proposed to detect rain streaks from all sub-views of the input LFI. Semi-supervised learning is introduced to accurately detect rain streaks by training on both virtual-world rainy LFIs and real-world rainy LFIs at multiple scales via calculating pseudo ground truth for real-world rain streaks. All sub-views subtracting the predicted rain streaks are then fed into a 4D residual model to estimate depth maps. Finally, all sub-views concatenated with the corresponding rain streaks and fog maps converted from the estimated depth maps are fed into a rainy LFI restoring model that is based on the adversarial recurrent neural network to progressively eliminate rain streaks and recover the rain-free LFI. Extensive quantitative and qualitative evaluations conducted on both synthetic LFIs and real-world LFIs demonstrate the effectiveness of our proposed method.


[41] 2208.07744

Secrecy Performance Analysis of RIS-aided Communication System with Randomly Flying Eavesdroppers

In this letter, we analyze the secrecy performance of a reconfigurable intelligent surface (RIS)-aided communication system with spatially random unmanned aerial vehicles (UAVs) acting as eavesdroppers. We consider the scenarios where the base station (BS) is equipped with single and multiple antennas.The signal-to-noise ratios (SNRs) of the legitimate user and the eavesdroppers are derived analytically and approximated through a computationally effective method. The ergodic secrecy capacity is approximated and derived in closed-form expressions.Simulation results validate the accuracy of the analytical and approximate expressions and show the security-enhanced effect of the deployment of the RIS.


[42] 2208.07754

Subtype-Aware Dynamic Unsupervised Domain Adaptation

Unsupervised domain adaptation (UDA) has been successfully applied to transfer knowledge from a labeled source domain to target domains without their labels. Recently introduced transferable prototypical networks (TPN) further addresses class-wise conditional alignment. In TPN, while the closeness of class centers between source and target domains is explicitly enforced in a latent space, the underlying fine-grained subtype structure and the cross-domain within-class compactness have not been fully investigated. To counter this, we propose a new approach to adaptively perform a fine-grained subtype-aware alignment to improve performance in the target domain without the subtype label in both domains. The insight of our approach is that the unlabeled subtypes in a class have the local proximity within a subtype, while exhibiting disparate characteristics, because of different conditional and label shifts. Specifically, we propose to simultaneously enforce subtype-wise compactness and class-wise separation, by utilizing intermediate pseudo-labels. In addition, we systematically investigate various scenarios with and without prior knowledge of subtype numbers, and propose to exploit the underlying subtype structure. Furthermore, a dynamic queue framework is developed to evolve the subtype cluster centroids steadily using an alternative processing scheme. Experimental results, carried out with multi-view congenital heart disease data and VisDA and DomainNet, show the effectiveness and validity of our subtype-aware UDA, compared with state-of-the-art UDA methods.


[43] 2208.07769

Unsupervised Domain Adaptation for Segmentation with Black-box Source Model

Unsupervised domain adaptation (UDA) has been widely used to transfer knowledge from a labeled source domain to an unlabeled target domain to counter the difficulty of labeling in a new domain. The training of conventional solutions usually relies on the existence of both source and target domain data. However, privacy of the large-scale and well-labeled data in the source domain and trained model parameters can become the major concern of cross center/domain collaborations. In this work, to address this, we propose a practical solution to UDA for segmentation with a black-box segmentation model trained in the source domain only, rather than original source data or a white-box source model. Specifically, we resort to a knowledge distillation scheme with exponential mixup decay (EMD) to gradually learn target-specific representations. In addition, unsupervised entropy minimization is further applied to regularization of the target domain confidence. We evaluated our framework on the BraTS 2018 database, achieving performance on par with white-box source model adaptation approaches.


[44] 2208.07797

On the Convergence of Inexact Gradient Descent with Controlled Synchronization Steps

We develop a gradient-like algorithm to minimize a sum of peer objective functions based on coordination through a peer interconnection network. The coordination admits two stages: the first is to constitute a gradient, possibly with errors, for updating locally replicated decision variables at each peer and the second is used for error-free averaging for synchronizing local replicas. Unlike many related algorithms, the errors permitted in our algorithm can cover a wide range of inexactnesses, as long as they are bounded. Moreover, the second stage is not conducted in a periodic manner, like many related algorithms. Instead, a locally verifiable criterion is devised to dynamically trigger the peer-to-peer coordination at the second stage, so that expensive communication overhead for error-free averaging can significantly be reduced. Finally, the convergence of the algorithm is established under mild conditions.


[45] 2208.07820

Deep Reinforcement Learning for RIS-aided Multiuser Full-Duplex Secure Communications with Hardware Impairments

In this paper, we investigate a reconfigurable intelligent surface (RIS)-aided multiuser full-duplex secure communication system with hardware impairments at transceivers and RIS, where multiple eavesdroppers overhear the two-way transmitted signals simultaneously, and an RIS is applied to enhance the secrecy performance. Aiming at maximizing the sum secrecy rate (SSR), a joint optimization problem of the transmit beamforming at the base station (BS) and the reflecting beamforming at the RIS is formulated under the transmit power constraint of the BS and the unit modulus constraint of the phase shifters. As the environment is time-varying and the system is high-dimensional, this non-convex optimization problem is mathematically intractable. A deep reinforcement learning (DRL)-based algorithm is explored to obtain the satisfactory solution by repeatedly interacting with and learning from the dynamic environment. Extensive simulation results illustrate that the DRL-based secure beamforming algorithm is proved to be significantly effective in improving the SSR. It is also found that the performance of the DRL-based method can be greatly improved and the convergence speed of neural network can be accelerated with appropriate neural network parameters.


[46] 2208.07824

A Deep Reinforcement Learning-based Adaptive Charging Policy for Wireless Rechargeable Sensor Networks

Wireless sensor networks consist of randomly distributed sensor nodes for monitoring targets or areas of interest. Maintaining the network for continuous surveillance is a challenge due to the limited battery capacity in each sensor. Wireless power transfer technology is emerging as a reliable solution for energizing the sensors by deploying a mobile charger (MC) to recharge the sensor. However, designing an optimal charging path for the MC is challenging because of uncertainties arising in the networks. The energy consumption rate of the sensors may fluctuate significantly due to unpredictable changes in the network topology, such as node failures. These changes also lead to shifts in the importance of each sensor, which are often assumed to be the same in existing works. We address these challenges in this paper by proposing a novel adaptive charging scheme using a deep reinforcement learning (DRL) approach. Specifically, we endow the MC with a charging policy that determines the next sensor to charge conditioning on the current state of the network. We then use a deep neural network to parametrize this charging policy, which will be trained by reinforcement learning techniques. Our model can adapt to spontaneous changes in the network topology. The empirical results show that the proposed algorithm outperforms the existing on-demand algorithms by a significant margin.


[47] 2208.07840

Multi-Pair D2D Communications Aided by An Active RIS over Spatially Correlated Channels with Phase Noise

This paper investigates a multi-pair device-to-device (D2D) communication system aided by an active reconfigurable intelligent surface (RIS) with phase noise and direct link. The approximate closed-form expression of the ergodic sum rate is derived over spatially correlated Rician fading channels with statistical channel state information (CSI). When the Rician factors go to infinity, the asymptotic expressions of the ergodic sum rates are presented to give insights in poor scattering environment. The power scaling law for the special case of a single D2D pair is presented without phase noise under uncorrelated Rician fading condition. Then, to solve the ergodic sum rate maximization problem, a method based on genetic algorithm (GA) is proposed for joint power control and discrete phase shifts optimization. Simulation results verify the accuracy of our derivations, and also show that the active RIS outperforms the passive RIS.