The rapid proliferation of delay-sensitive Internet of Things (IoT) applications in 6G networks necessitates innovative solutions for computation-intensive tasks in mobile edge computing (MEC) systems. This paper addresses the critical challenge of spatio-temporal task heterogeneity induced by user mobility in multi-server MEC-enabled IoT networks. A cooperative MEC framework is proposed with joint optimization of transmit power allocation, computation offloading, and central processing unit (CPU)-cycle frequencies. The formulated mixed-integer nonlinear programming problem is decomposed into two tractable subproblems: the transmit power control subproblem solved by LambertWfunction-based closed-form solutions; the task offloading and CPU-cycle frequency allocation subproblem solved by a novel deep reinforcement learning (DRL)-based method named D3SAC. Combining dueling double deep Q-network with soft actor-critic algorithms, our proposed D3SAC method dynamically adapts to user mobility while maintaining energy efficiency and computational latency. Simulation results demonstrate over 52% energy reduction compared with maximum-power transmission schemes. Compared to existing DRL baselines, our proposed method achieves superior task completion rate and CPU-cycle frequencies utilization efficiency while maintaining balanced workload distribution.

Blind Image Quality Assessment faces challenges in enabling computational models to mimic the hierarchical progressive perception mechanisms of the Human Visual System (HVS). Existing methods often neglect the two-stage process of HVS—global distortion identification followed by local quality evaluation—and its distinct sensitivity to distortion types. To address this, we propose Fre-QNet, a novel framework integrating two key components: (1) A Quaternion Progressive Perception (QPP) module that hierarchically extracts multi-scale spatial features using quaternion convolution, explicitly simulating the global-to-local observation process of HVS while enhancing cross-scale interactions; (2) A Frequency Prompting (FP) module that quantifies distortion types and severity in the Fourier domain by leveraging frequency patterns of common distortions and the sensitivity variations of HVS. The QPP and FP modules collaboratively embed biological vision principles into computational modeling through dual-domain feature learning, with the QPP module directly anchoring the core logic of progressive perception. Experiments on TID2013 and CSIQ benchmarks demonstrate Fre-QNet’s superiority over state-of-the-art methods, validating its effectiveness in matching human perceptual quality judgments. Our source code is available at: https://github.com/hhsda/Fre-QNet .

No-reference image quality assessment (NR-IQA) plays a pivotal role in computer vision by enabling image quality evaluation without reference images. While recent CNN and Transformer-based methods have advanced feature extraction, they face significant limitations. CNNs exhibit local feature bias, limiting their ability to capture global dependencies and complex structures critical to understanding diverse distortions. Transformers, despite modeling nonlocal dependencies through multihead attention, suffer from quadratic computational complexity with spatial dimensions, hindering efficient multiscale analysis. Moreover, their attention mechanisms frequently overlook critical interchannel dependencies, which are vital for capturing fine details in texture-rich images. Coupled with difficulties in handling high-noise environments and complex textures, this results in limited real-world accuracy and poor generalization across diverse datasets and unknown distortions. To bridge these gaps, we propose WAQNIQA, a novel wavelet-augmented quaternion network for NR-IQA. Distinct from conventional architectures, WAQNIQA integrates two synergistic modules: the wavelet-infused adaptive attention (WIAA) module, which leverages wavelet transforms (WTs) to achieve robust multiscale spatial-frequency analysis with linear complexity, and the quaternion collaborative feature enhancement (QCFE) module, which holistically models interchannel correlations to preserve fine texture details. Furthermore, we introduce PowerGridIQ, the first NR-IQA dataset specifically tailored for power grid scenarios. Extensive experiments demonstrate that WAQNIQA consistently surpasses state-of-the-art CNN and Transformer-based methods on PowerGridIQ and six public benchmarks. Notably, WAQNIQA exhibits superior cross-domain generalization, achieving competitive performance on the AGIQA-1K dataset for AI-generated content (AIGC) without explicit semantic alignment training, thereby validating its robustness against diverse and unknown distortions. Our code is available athttps://github.com/king-huoye/WAQNIQA

The precise identification of key landmarks on three-dimensional tooth mesh models is paramount for computer-aided orthodontic treatment. However, existing methodologies exhibit limitations with respect to the integration of global and local features, which undermines accuracy in complex scenarios and excessively emphasizes relative landmark positions, resulting in displacement errors. To mitigate these issues, this study introduces DGLL, a hybrid feature learning network characterized by a dual-branch architecture that amalgamates global and local features. DGLL integrates a Cascaded Topological Relation Module (CTRM) to stabilize the extraction of global features and a Pan-scale Feature Modulation Module (PFMM) to balance relative and absolute positional accuracy. Empirical evaluations across various tooth types demonstrate that DGLL consistently enhances the accuracy of landmark localization. This research provides an effective approach to the automated analysis of tooth data, thereby improving the precision and efficacy of orthodontic treatment.

Accurate segmentation and tracking of cells are crucial in biomedical research. The latest methods utilize multilayer convolution or block based region similarity measurement to process foreground background relationships, but under low signal-to-noise ratio conditions, there is insufficient target pattern recognition and boundary detail characterization, which can easily lead to segmentation errors and boundary blurring. To address this issue, we propose SimCNet, a segmentation network based on similarity measurement and texture enhancement. The network consists of three modules: the Global Pattern Sorting and Focusing Module (GPRF), which uses pixel level feature histogram aggregation and a Fourier based feedforward network to create similarity fingerprints for different target patterns, enhancing the recognition ability of different regions. The Multi Scale Feature Collaborative Capture Module (MFCM) enhances the global and inter channel information exchange of GPRF through the fusion of multi-scale fusion mechanisms, eliminating semantic gaps. Texture Feature Enhancement Module (TFEM), which uses quaternion convolution to enrich texture representations in similarity feature maps and enhance boundary perception. Experiments have shown that SimCNet exhibits excellent performance and generalization on multiple publicly available datasets such as MoNuSeg.

Accurate segmentation of brain tumor regions from multi-modal magnetic resonance imaging (MRI) is critical for clinical diagnosis. However, missing modalities is a common issue in clinical practice, where the unavailability of certain imaging modalities complicates the extraction and integration of complementary information across multiple modalities, leading to a decline in segmentation accuracy. Many existing models fail to adequately address subtle structural changes and boundary information within tumor regions when faced with missing modalities, limiting their ability to effectively adapt to complex tumor morphologies. To tackle these issues, The Multi-Scale Adaptively-Aware and Recalibration Network (MARNet) proposed in this paper can adaptively and fully explore the potential of multi-modal data under different combinations in the presence of missing modalities. MARNet incorporates a Feature Recalibration and Enhancement Module (FREM) that recalibrates and enhances the three-dimensional feature representation, emphasizing important fine-grained features of brain tumors. Subsequently, the Adaptive Shape-Aware Fusion Module (ASFM) fully exploits available modality information, achieving adaptive feature fusion for varying tumor locations and shapes, thereby compensating for information loss due to missing modalities. Furthermore, the Global and Multiscale Feature Integration Module (GMFIM) is designed to effectively capture long-range dependencies of tumors, particularly under conditions of missing modalities, aiding in the restoration and reconstruction of complete tumor structures. Extensive experiments on the BraTS2020, BraTS2018 and BraTS2015 datasets demonstrate that the proposed method surpasses several advanced brain tumor segmentation approaches in the context of missing modalities.

Current remote sensing image dehazing models often encounter challenges under dense haze conditions due to the significant loss of high-frequency information, impairing accurate scene recovery. In addition, these models typically do not incorporate additional sensor information to guide the generation of dehazed images. However, satellites generally house a variety of image sensors, among which panchromatic sensors are included. Panchromatic (PAN) images, which usually present clear boundary information, could potentially assist models in generating dehazed images. We introduce Guidance Net, an innovative model that utilizes historical PAN images to enhance the dehazing process. Specifically, we introduce the Adaptive Self-Attention Interest Texture Filter (ASAITF) for the effective integration of guidance information from PAN images. Additionally, recognizing the limitations of existing methods that predominantly emphasize low-frequency features, we propose the Redundancy Filtering Mechanism (RFM), aimed at efficient high-frequency feature extraction and seamless integration within Vision Transformer architectures. To ensure a comprehensive evaluation, we also present the PAN Guidance dataset. Experimental results indicate that Guidance-Net surpasses state-of-the-art methods in generating dehazed images guided by prompts.

Few-shot Knowledge Graph Entity Typing (KGET) is important in current knowledge graph research. KGET is a task aimed at predicting the missing types of entities in a knowledge graph. Previous methods require available and sufficient training instances to achieve satisfactory results, but the long-tail distribution problem makes existing methods based on deep learning ineffective in the real world. Moreover, the complete combination of neighbors, including relations, entities and their types, along with their interactions, can significantly enhance type inference, yet have not been fully explored. Therefore, we propose a novel model named Few-Shot entity typing with Enhanced Neighbor (FSEN) for few-shot knowledge graph entity typing, which contains the Critical Neighbor Aggregation Module to learn the interaction among complete neighbors, and the Adaptive ProtoNet for few-shot entity typing. Specifically, in Critical Neighbor Aggregation Module, effective combinations of relation, entity and type are selected, and central entities are enhanced with more premium and complete neighbor information. In Adaptive ProtoNet, the position of the prototype in prototype space is adaptively adjusted to achieve a more balanced distribution, thus improving the inference performance under few-shot setting. The results of the experiment on two public datasets show that our method improves the performance of MRR by at least 5.2% compared with the existing method under 5-shot conditions.

Wireless Power Transfer technology, which can charge IoT devices over the air, has become a promising technology for IoT networks. In wireless-powered IoT networks, broadcasting is a fundamental networking service for disseminating messages to the whole network. To seek a fast and collision-free broadcast schedule, the problem of Minimum Latency Broadcast Scheduling (MLBS) has been well studied when nodes are energy-abundant. However, in wireless-powered networks, a node can only receive or transmit packets after it has harvested enough energy. In such networks, it is of great importance to exploit the divergent harvested energy to reduce the broadcast latency. Unfortunately, existing works always assume a predetermined tree and a fixed transmission power for broadcast scheduling, which greatly limits their performance. Thus, in this article, we investigate the first work for the MLBS problem in wireless-powered networks without relying on predetermined trees. First, the problem is formulated and proved to be NP-hard. Then, two structure-adaptive scheduling algorithms are proposed with a theoretical bound, which can intertwine the construction of broadcast tree with the computation of an energy-aware schedule simultaneously. Furthermore, a power-aware scheduling method is also proposed to take the structure of the broadcast tree, the adjustment of nodes’ transmission powers, and the interference during transmissions into account simultaneously. Additionally, the algorithm for the MLBS problem under the physical interference model is also studied. Finally, the theoretical analysis and simulation results verify that the proposed algorithms have high performance in terms of latency.

Accurate prediction of mortality in nasopharyngeal carcinoma (NPC), a complex malignancy particularly challenging in advanced stages, is crucial for optimizing treatment strategies and improving patient outcomes. However, this predictive process is often compromised by the high-dimensional and heterogeneous nature of NPC-related data, coupled with the pervasive issue of incomplete multi-modal data, manifesting as missing radiological images or incomplete diagnostic reports. Traditional machine learning approaches suffer significant performance degradation when faced with such incomplete data, as they fail to effectively handle the high-dimensionality and intricate correlations across modalities. Even advanced multi-modal learning techniques like Transformers struggle to maintain robust performance in the presence of missing modalities, as they lack specialized mechanisms to adaptively integrate and align the diverse data types, while also capturing nuanced patterns and contextual relationships within the complex NPC data. To address these problem, we introduce IMAN: an adaptive network for robust NPC mortality prediction with missing modalities. IMAN features three integrated modules: the Dynamic Cross-Modal Calibration (DCMC) module employs adaptive, learnable parameters to scale and align medical images and field data; the Spatial-Contextual Attention Integration (SCAI) module enhances traditional Transformers by incorporating positional information within the self-attention mechanism, improving multi-modal feature integration; and the Context-Aware Feature Acquisition (CAFA) module adjusts convolution kernel positions through learnable offsets, allowing for adaptive feature capture across various scales and orientations in medical image modalities. Extensive experiments on our proprietary NPC dataset demonstrate IMAN’s robustness and high predictive accuracy, even with missing data. Compared to existing methods, IMAN consistently outperforms in scenarios with incomplete data, representing a significant advancement in mortality prediction for medical diagnostics and treatment planning. Our code is available at https://github.com/king-huoye/BIBM-2024/tree/master.

Sequential Recommendation (SR) aims to understand user preferences by analyzing historical interactions with items. Recent approaches have shifted from the time domain to the frequency domain to potentially enhance preference modeling. While fast Fourier transform is a common choice for frequency transform, it may introduce issues like the Gibbs phenomenon, leading to potentially suboptimal model performance. To address this, we introduce discrete cosine transform into sequential recommendation and present a novel multi-frequency transformation sequential recommendation, named HMFTRec, within a hierarchical framework. Specifically, we develop a discrete cosine transform module base on channel attention. A hierarchical spectrum framework that combines Fourier and discrete cosine transforms is introduced to capture finer-grained frequency domain information and mitigate the Gibbs phenomenon to some extent. Furthermore, contrastive learning is employed to potentially enhance the quality of user embeddings learned from the frequency domain. Extensive experiments conducted on four widely recognized benchmark datasets demonstrate that our model significantly outperforms state-of-the-art approaches.

Nested Named Entity Recognition (NNER) focuses on addressing overlapped entity recognition. Compared to Flat Named Entity Recognition (FNER), annotated resources are scarce in the corpus for NNER. Data augmentation is an effective approach to address the insufficient annotated corpus. However, there is a significant lack of exploration in data augmentation methods for NNER. Due to the presence of nested entities in NNER, existing data augmentation methods cannot be directly applied to NNER tasks. Therefore, in this work, we focus on data augmentation for NNER and resort to more expressive structures, Composited-Nested-Label Classification (CNLC) in which constituents are combined by nested-word and nested-label, to model nested entities. The dataset is augmented using the Composited-Nested-Learning (CNL). In addition, we propose the Confidence Filtering Mechanism (CFM) for a more efficient selection of generated data. Experimental results demonstrate that this approach results in improvements in ACE2004 and ACE2005 and alleviates the impact of sample imbalance.

Distantly supervised relation extraction (DSRE) aims to automatically identify relation facts from unstructured text. Most current DSRE works solve the noise problem based on the bag-level, but the denoising ability of these methods decreases when the bag consists of fewer sentences. In this study, we propose a Distantly supervised Relation extraction with Label Prompt (DRLP) framework. We use textual labels (such as label names) as label prompts to alleviate the problem of decreased denoising ability by utilizing the information of entities and relations in label names. During the training process, label prompts are directly connected to the sentences in the bag to provide a more comprehensive bag representation, and label prompts are randomly deleted based on the number of sentences in the bag. Moreover, we design a residual selective attention mechanism that minimizes the influence of spurious features and optimizes the utilization of label information. Our framework is evaluated on NYT-10d and NYT-10m, the results indicate that our method outperforms the state-of-the-art methods.

In recent years, many Lesion segmentation (LS) models based on UNet have been proposed. However, existing researches rarely consider the influence of illumination change leads to the weak boundary area. Such as melanomas and polyps, the demarcation of the boundary between the diseased area and the surrounding tissue remains particularly challenging. To overcome these challenges, we propose an IlluminationAdaptive Multi-scale Lesion Segmentation Network (IAMS-Net). In IAMS-Net, we integrate Illumination-Adaptive MultiStream Attention (IAMA) and Contour Perception Module (CPM). In the decoding stage, the IAMA is used as a bridge between the encoder and the decoder to solve the adverse effects of illumination changes on the segmentation of weak boundary lesions. In order to further enhance the boundary features lost due to illumination change in the low-contrast lesion area, we introduce the CPM to improve the perception of the integrity of the lesion area. Subsequently, we performed comparison and ablation experiments using the publicly available ISIC2018 dataset and the individually collected data set BoreIllumination(BI).

Summary Data broadcast is a fundamental communication pattern in wireless IoT networks, in which the messages are disseminated from a source node to the entire network. The problem of minimum latency broadcast scheduling (MLBS) which is aimed to generate a quick and conflict‐free broadcast schedule has not been extensively explored in duty‐cycled networks. The existing works either work in a centralized scheme or rely on a fixed tree for broadcasting. Additionally, they all employ a strict premise that each node can only utilize one channel for both transmitting and receiving messages. Thus, to address the issues mentioned above, we examine the first distributed broadcasting algorithm in multi‐channel duty‐cycled wireless IoT networks, without relying on a predetermined tree. First, the MLBS problem in such networks is defined and proved to be NP‐hard. Then, in order to avoid transmission conflicts between different links locally, two efficient data structures are designed to help compute the earliest time and channel of receiving messages without conflicts. Based on the above data structures, we introduce an efficient distributed broadcasting algorithm, which can generate a latency‐sensitive broadcast tree while calculating a collision‐free broadcast schedule, simultaneously. Finally, the theoretical analysis and simulations demonstrate the efficiency of the proposed algorithm.

More and more people around the world suffer from insomnia, hence sleep-aid methods are very important and urgent. In this article, a flexible sleep-aid eye mask based on acupoint electric pulse stimulation (AEPS) combined bioelectrical signal feedback is proposed. Besides, the massage sleep-aid principle of Yintang and Anmian acupoints is described. Afterwards, AEPS circuit, bioelectrical signal acquisition (BSA) circuit, and a smart flexible sleep-aid eye mask are put forward. The sleep-aid effect evaluation methods based on the key features of EEG and ECG signals, as well as the real-time sleep-aid control method, are presented in detail. Experimental results demonstrate that the BSA circuit and feature extraction method with discrete wavelet transform (DWT) analysis are effective. After four kinds of experimental tests with ten subjects, compared with the results of no stimulation (NS), the values of HRMEAN and$E_{\mathrm{ betam}}$of AEPS of Yintang acupoint (AEPS-Y), AEPS of Anmian acupoint (AEPS-A), and AEPS of Yintang and Anmian acupoints (AEPS-YA) decrease, while the values of RRMEAN, SDNN, RMSSD, LF, HF, RLH,$E_{\mathrm{ alpham}}$, and RAB of AEPS-Y, AEPS-A, and AEPS-YA increase, which verifies that the sleep-aid controls of AEPS-Y, AEPS-A, and AEPS-YA are all effective. In addition, the sleep-aid effect of AEPS-Y is better than that of AEPS-A or AEPS-YA. Compared with that without AEPS-Y, the median sleep latency with AEPS-Y of the flexible eye mask is reduced by about 43.08%. To sum up, this flexible eye mask can be widely used for sleep-aid control at home, which is low cost and comfortable.

With the wide application of the Internet of Things technology, the hardware security has attracted more and more attention from users around the world. Hardware Trojan (HT) of integrated circuit (IC) has become a main security threat gradually. Therefore, HT detection is very significant. In this article, a HT automatic test system used for side-channel test combined logic test is constructed with a high-performance oscilloscope, FPGA chips, a NI digital acquisition card and LabVIEW software. Besides, the test flow chart and data processing method are depicted in detail. Spectral feature analysis combined principal component analysis is proposed for feature extraction. Fuzzy C-means clustering combined spectral energy analysis is put forward to distinguish the Trojan category from the golden category. Then Fusion distance (i.e. Mahalanobis distance combined Euclidean distance) is presented for the real-time HT recognition. A 128-bit AES cipher circuit and a 2-bit counter are applied as a golden circuit and a Trojan circuit, respectively. Experimental results demonstrate that the detection accuracy is 100% and the proposed detection method can easily achieve 0.1% HT detection sensitivity, which verifies that the detection method is feasible and effective.

Emotions are complex, and people vary greatly in their accuracy in recognizing their own emotions and those of others. With advances in computer science and neuroscience, there is a desire to use automated techniques to help people identify emotions. Bio-electrical signals have been proven effective for emotion detection, but the acquisition of conventional electrocardiogram (ECG) and EEG requires medical-specific equipment, which is very expensive, uncomfortable, and inconvenient due to the large number of electrodes and the hair-covered scalp. In this article, a novel emotion recognition method based on the feature fusion of single-lead EEG and ECG signals is proposed, using the long short term memory (LSTM)-MLP-based model and the CNN-based model for feature fusion and classification, respectively, with fivefold cross-validation for validation. The ECG and EEG signals of 15 participants were collected in five states: 1) happy; 2) relaxed; 3) calm; 4) sad; and 5) afraid, each of which was stimulated using the participants’ own proposed music. Various time-domain features, frequency-domain features, and nonlinear features were extracted from the ECG and EEG signals. Experimental results demonstrate that the accuracy of emotion recognition and classification of signals captured by the proposed device can reach 92.08% using the CNN model. While using the LSTM-MLP feature fusion model, the accuracy figure can be improved to 95.07%. The results of the ablation experiment indicate that the feature fusion approach does improve the accuracy of recognition. It is demonstrated that the proposed device and emotional recognition approach are effective and feasible.

The solution to boundary value problems is of great significance in industrial software applications. In this paper, we propose a novel deep learning method for simulating stress field distributions in simply supported beams, aiming to serve as a solver for stress boundary value problems. Our regression network, Stress-EA, utilizes the convolution encoder module and additive attention to accurately estimate the stress in the beam. By comparing the Stress-EA prediction results with the stress values calculated using ABAQUS, we achieve a mean absolute error (MAE) of less than 0.06. This indicates a high level of consistency between the stress values obtained from the two approaches. Moreover, the prediction time of Stress-EA is significantly shorter, taking only 0.0011s, compared to the calculation time of ABAQUS, which is 16.91s. This demonstrates the high accuracy and low computational latency of our model. Furthermore, our model exhibits smaller model parameters, requires less computation, and has a shorter prediction time compared to training results obtained using classic and advanced networks. To accelerate training, we utilize data parallel methods, achieving up to 1.89 speedup on a dual-GPU platform without compromising accuracy. This advancement enhances the computing efficiency for large-scale industrial software applications.

Patch-level histological tissue classification is an effective pre-processing method for histological slide analysis. However, the classification of tissue with deep learning requires expensive annotation costs. To alleviate the limitations of annotation budgets, the application of active learning (AL) to histological tissue classification is a promising solution. Nevertheless, there is a large imbalance in performance between categories during application, and the tissue corresponding to the categories with relatively insufficient performance are equally important for cancer diagnosis. In this paper, we propose an active learning framework called ICAL, which contains Incorrectness Negative Pre-training (INP) and Category-wise Curriculum Querying (CCQ) to address the above problem from the perspective of category-to-category and from the perspective of categories themselves, respectively. In particular, INP incorporates the unique mechanism of active learning to treat the incorrect prediction results that obtained from CCQ as complementary labels for negative pre-training, in order to better distinguish similar categories during the training process. CCQ adjusts the query weights based on the learning status on each category by the model trained by INP, and utilizes uncertainty to evaluate and compensate for query bias caused by inadequate category performance. Experimental results on two histological tissue classification datasets demonstrate that ICAL achieves performance approaching that of fully supervised learning with less than 16% of the labeled data. In comparison to the state-of-the-art active learning algorithms, ICAL achieved better and more balanced performance in all categories and maintained robustness with extremely low annotation budgets. The source code will be released at https://github.com/LactorHwt/ICAL.

The rapid development of artificial intelligence provides rich technologies and tools for the automated understanding of literary works. As a comprehensive carrier of storylines, movies are natural multimodal data sources that provide sufficient data foundations, and how to fully leverage the benefits of data remains a sustainable research hotspot. In addition, the efficient representation of multi-source data also poses new challenges for information fusion technology. Therefore, we propose a knowledge-enhanced turning points identification (KTPi) method for multimodal scene recognition. First, the BiLSTM method is used to encode scene text and integrate contextual information into scene representations to complete text sequence modeling. Then, the graph structure is used to model all scenes, which strengthens long-range semantic dependencies between scenes and enhances scene representations using graph convolution network. After, the self-supervised method is used to obtain the optimal number of neighboring nodes in sparse graph. Next, actor and verb knowledge involved in the scene text are added to the multimodal data to enhance the diversity of scene feature expressions. Finally, the teacher-student network strategy is used to train the KTPi model. Experimental results show that KTPi outperforms baseline methods in scene role recognition tasks, and ablation experiments show that incorporating knowledge into multimodal model can improve its performance.

Multicast is a fundamental function in network traffic engineering, allowing data traffic to be delivered from the source node to multiple destinations efficiently. To ensure the reliability and security of data traffic, NFV-enabled multicast (Network Function Virtualization) has emerged as a promising technology to reduce deployment and maintenance costs in mobile edge clouds, and has drawn extensive researching interests recently. However, existing works all assume that the Service Function Chain (SFC) follows a fixed-order, which greatly limits its application. Therefore, in this paper, we propose the first work to address the sequential SFC embedding problem without a fixed order for NFV-enabled multicasting in mobile edge clouds. Firstly, we formulate such a minimum cost SFC embedding problem and prove it to be NP-hard. Secondly, we propose a min-path breadth-first based progressive embedding algorithm (MBPE) for NFV-enabled multicasting, which achieves an approximation ratio of 1+K, where K represents the approximation ratio of the Steiner tree problem. Finally, the experiments demonstrate the high efficiency of the proposed method compared to the state-of-the-art algorithms.

The capabilities of deep models are constantly mined for extraction and representation of features among text classification tasks. However, these models are sensitive to changes in input data, resulting in poor robustness. Meanwhile, the model lacks information interaction and weak representation ability. In this work, for feature extraction, a joint model that consists of a convolutional neural network, a bidirectional gated recurrent unit, and an attention mechanism is proposed. This new model can improve versatility and fully discover category information in text. For feature representation, a projector under the supervised contrastive learning method is introduced. The method can improve the representation of an encoder and realize aggregation of the same category. Considering the robustness of the PCRA, the gradient penalty is added to a contrastive loss function. Experiments are performed on four datasets to assess the proposed model (PCRA and PCRA‐GP) using an accuracy metric. The experimental results show that our model is suitable for variable‐length and bilingual texts. Compared with the baseline model, it remains competitive, and it reaches SOTA on the 20 Newsgroups dataset. Moreover, the performance of the model is evaluated under different hyperparameters to clarify its working mechanism.

With the rapid development of economy, science, and technology, the aging issues become more and more serious. People aged above 65 have a risk of 28%–35% to fall. Among them, bedside falls happen most frequently. Therefore, the capability to detect fall events of the elderly is very important. In this article, a novel noncontact fall detector based on a MEMS low-resolution infrared sensor and a low-cost radar sensor is developed to detect bedside fall. Besides, IR image processing algorithms based on the adaptive filter, successive approximation, double boundary scans, and mathematical morphology processing are proposed in detail. Partition processing algorithm is used to suppress the influence of residual or existed heat sources on the bed or ground. Then, the statistical features of the center, area, temperature and duration, as well as stable flag and fall action flag, are extracted for fall recognition. Finally, a three-layer radial basis function neural network is applied to distinguish the fall events from the nonfall events. Considering the influence factors of ambient temperature, brightness, gender, dressing, fall posture, fall location, and scenario, a total of 640 tests are conducted and 5-fold cross validation is used to evaluate the classification performance. Experimental results indicate that the averages of the recall, precision, F1-Score, and detection accuracy are measured to be 91.25%, 94.76%, 92.97%, and 93.13%, respectively, which demonstrates that the proposed fall detection method is effective. Besides, the detection accuracy decreases from 96.88% to 85.94% as the ambient temperature rises. Hence, this noncontact fall detector can be widely applied for bedside fall detection at home, which is low cost, nonwearable, unobtrusive, noninvasive, and privacy preserved.

Massive GPU acceleration processors have been used in high-performance computing systems. The Dennard scaling has led to power and thermal constraints limiting the performance of such systems. The demand for both increased performance and energy efficiency is highly desired. This article presents a multilayer low-power optimization method for warps and tasks parallelisms. We present a dynamic frequency regulation scheme for performance parameters in terms of load balance and load imbalance. The method monitors the energy parameters in runtime and adjusts adaptively the voltage level to ensure performance efficiency with energy reduction. The experimental results show that the multilayer low-power optimization with dynamic frequency regulation can achieve 40% energy consumption reduction with only 1.6% performance degradation, thus reducing 59% maximum energy consumption. It can further save about 30% energy consumption in comparison with the single-layer energy optimization.

Age of Information (AoI) has emerged as a new metric to measure data freshness from the destination’s perspective. The problem of optimizing AoI has been attracting extensive interests recently. However, existing works mainly focused on scheduling data transmission for AoI optimization. While at wireless-powered network edge, the charging plan of source nodes also requires to be computed in advance, which means the system AoI is determined by not only the data transmission decision but also the charging plan. Thus, in this paper, we investigate the first work to optimize the weighted peak AoI from the point of charging at wireless-powered network edge with a directional charger. Firstly, to minimize the weighted sum of average peak AoI, the AoI minimization problem is transformed to a charging time optimization problem with respect to the overlapped charging areas and average peak AoI, and an approximate algorithm is proposed to obtain the required charging time for each source node. Then, an age-based scheduling algorithm is proposed to compute the charging and data transmission decisions for each source node simultaneously, which can not only optimize the weighted sum of average peak AoI but also guarantee the maximum peak AoI for each source node. The proposed algorithm is proved to have an approximation ratio of up to (1+φ), where φ is a much smaller value related to the weight of each source node. Finally, the simulation results verify the high performance of proposed algorithms in terms of average and maximum peak AoI.

Named entity recognition (NER) is one of the basic tasks of knowledge extraction. In view of corpus in some specific fields have sparse semantics and limited text standardization, the existing entity recognition methods have the problems of wrong potential word interference, matching potential words in some specific fields difficultly, and the key information of entities with different lengths in sentences may have interference in the representation of attention mechanism. This paper proposes a NER model integrating word and improved entity key information attention mechanism, which is called binocular attention-based BiLSTM with CNN network (BACBN). Firstly, the character feature embedding is generated through bidirectional encoder representation of transformers (BERT), and a word-level character feature attention mechanism is proposed to highlight the character features constituting words. Secondly, based on the bidirectional long short-term memory network (BiLSTM), an n-gram pooling feature attention mechanism is proposed. The prominent features corresponding to different convolution kernel sizes are obtained through convolution neural network (CNN), and the context features are weighted according to the prominent features, so as to obtain the information that contributes more to entity recognition with different lengths. The experimental results on four specific field Chinese corpus NER datasets show that BACBN model can improve the result of entity recognition.

Taking fault diagnosis corpus as the research object, an event logic knowledge graph construction method is proposed in this paper. Firstly, we propose a data labeling strategy based on a constructed event logic ontology model, then collect large-scale robot transmission system fault diagnosis corpus, and label part of the data according to the strategy. Secondly, we propose a transfer learning model called Trans-SBLGCN for event argument entity and event argument relation joint extraction. A language model is trained based on large-scale unlabeled fault diagnosis corpus and transferred to a model based on stacked bidirectional long short term memory (BiLSTM) and bidirectional graph convolutional network (BiGCN). Experimental results show that the method is superior to other methods. Finally, an event logic knowledge graph of robot transmission system fault diagnosis is constructed to provide decision support for autonomous robot transmission system fault diagnosis.

Age of Information (AoI), emerged as a new metric to quantify the data freshness, has attracted increasing interests recently. Most existing works try to optimize the system AoI from the point of data transmission. Unfortunately, at wireless-powered network edge, the charging schedule of the source nodes also needs to be decided besides data transmission. Thus, in this paper, we investigate the joint scheduling problem of data transmission and energy replenishment to optimize the peak AoI at network edge with directional chargers. To the best of our knowledge, this is the first work that considers such two problems simultaneously. Firstly, the theoretical bounds of the peak AoI with respect to the charging latency are derived. Secondly, for the minimum peak AoI scheduling problem with a single charger, an optimal scheduling algorithm is proposed to minimize the charging latency, and then a data transmission scheduling strategy is also given to optimize the peak AoI. The proposed algorithm is proved to have a constant approximation ratio of up to 1.5. When there exist multiple chargers, an approximate algorithm is also proposed to minimize the charging latency and peak AoI. Finally, the simulation results verify the high performance of proposed algorithms in terms of AoI.

Summary The Internet of Things is becoming widely popular in the past decade, which comes with huge amount of data. These magnanimous data, stored in data centers, put forward the new demand for the efficient management of the network. In this article, we propose Software‐Defined Congestion Control Plane (SDCCP), a hybrid network control architecture that aims to fully utilize the network while avoiding congestion. SDCCP is based on Software‐Defined Networking and CCP, in which the controller collects the network statistics and specifies the behavior of the end‐to‐end hosts by sending feedback or modifying their transport layer parameters directly. It can also be used to mitigate Distributed Denial of Service attacks and other security problems. In addition, we propose FCA, a Feedback‐based Congestion Avoidance algorithm running on SDCCP, which adapts the congestion window based on the feedback from the remote controller. We evaluate SDCCP and FCA in Mininet and the result shows that FCA can achieve high network utilization while keeping the queue length of the routers in a low level. Also, FCA is robust to noncongestion loss, and outperforms other algorithms at high loss rate.

Abstract Single infrared (IR) image super‐resolution methods can help to reduce the cost and difficulty in manufacturing IR sensors for the imaging system. However, the deep learning‐based image SR methods need to build a complex network and thus consume a lot of computational power, which limits the application of SR technology on devices with low computing resources in practice. To solve this problem, the authors present a lightweight multi‐path feature fusion network (MFFN) for the single infrared (IR) image SR. A multi‐path feature fusion block (MFFB) is developed to extract and fuse multiple and discriminative features in a recursive feedback way. Specifically, the multiple features are refined via the linear feature extraction branch, shared‐source residual feature extraction branch, and channel attention branch in MFFB. Finally, the authors reconstruct the high‐resolution IR images from the low‐resolution counterpart based on the refined multiple features. The experimental results demonstrate that MFFN achieves high‐quality single infrared image SR and shows superiority over previous methods for several scale factors (e.g. ×2, ×3, and ×4). MFFN has potential applications in the mobile infrared imaging system.

People spend about one third of their lifetime in sleep, and sleep quality has a great impact on people’s health. Therefore, vital signs and sleep quality monitoring are more and more important. In this article, a novel smart flexible sleep monitoring belt with MEMS triaxial accelerometer and pressure sensor is developed to detect vital signs, snore events, and sleep stages. Besides, the related algorithms and methods for data preprocessing, heart and respiration rates detection, snoring recognition, and sleep stages classification are proposed in detail. Then, a series of sleep experiments is performed based on the experimental platform for the smart flexible belt, and the test results measured by PolySomnoGraphy are used as the golden standards for comparison. The experimental results demonstrate that the detection accuracies of heart rate and respiration rate of the belt are about 1.5 and 0.7 bpm, respectively. The accuracy of 97.2% is achieved by the proposed snoring recognition method. In addition, as for 2-stage analysis, the sensitivities of awake and asleep stages are 90.2% and 100%, respectively; meanwhile, the corresponding accuracy of sleep stages prediction is as high as 95.1%. However, as for 4-stage analysis, the sensitivities of awake, rapid eye movement, light sleep, and deep sleep stages are 90.2%, 77.1%, 78.1%, and 73.5%, respectively; meanwhile, the corresponding accuracy of sleep stages prediction decreases to 79.7%. In general, the test results indicate that vital signs detection, snoring recognition, and sleep stages classification based on the sleep monitoring belt are feasible and effective. Hence, this smart flexible belt can be widely used for sleep monitoring at home due to low cost and high performance.

Sleep apnea impacts more and more people all over the world, and obstructive sleep apnea of which is the most frequent. Hence, research on snoring detection and related suppression methods is extremely urgent. In this article, a novel low-cost flexible patch with MEMS microphone and accelerometer is developed to detect snore event and sleeping posture, and a small vibration motor embedded in the patch is designed to suppress snoring. Theoretical analyses of short-time energy, piecewise average filtering (PAF), and Mel-frequency cepstral coefficients (MFCCs) processing are described in detail, and the improved MFCCs are put forward and used as the input of the convolutional neural network (CNN). Furthermore, the snore recognition method based on the combination of similarity analysis and CNN analysis is presented, followed by the snoring suppression method. Experimental results demonstrate that the main features of the sound signals can be extracted effectively by PAF and MFCCs processing, and the data compression ratio is about 99.41%. Besides, the locations of the eigenvectors can be found accurately based on short-time energy analysis. The numbers of high similarity of snoring signals within 30 s are larger than 3, while those of non-snoring signals are often less than 3. If the preliminary screening with similarity analysis is passed, CNN analysis will be conducted to judge whether there are snoring events. The accuracy of snore recognition with CNN analysis is calculated to be as high as 99.25%. Finally, the average snoring time measured by the smart patch with snoring suppression is reduced to 15 from 135 min, which indicates that the proposed snore recognition and suppression methods are effective.

Monitoring crop growth is of great significance to obtain crop growth status information for development of smart agriculture. The traditional way to measure the phenotypic parameters of crops is labor-intensive and encounters inconvenient operations. In this study, we propose to obtain the phenotypic parameters of crops from 3-D reconstruction of plants from single RGB images using a data-driven plant phenotypic parameters estimation network (P3ES-Net) deep neural network, which enables to estimate the depth shift and camera focal length used for depth estimation and reconstruction of the 3-D model of plants. Based on the principles of the monocular ranging and pinhole imaging model, crop phenotypic parameters such as height, canopy size, and trunk diameter can then be calculated from the 3-D model. Experiments with four practical plants present that our method is able to achieve acceptable evaluation of the growth status of plants. Of more significance, it achieves particular superior depth estimation performance over a commercial depth camera, which is a very new on-sale depth camera using stereo vision and deep learning network. This potential performance throws light on the low-cost measurement of crop phenotypic parameters using RGB camera in monitoring crop growth.

With advances in wireless power transfer techniques, battery-free wireless sensor networks (BF-WSNs) which can support long-term applications, has been attracting increasing interests in recent years. Unfortunately, the problem of minimum latency aggregation scheduling (MLAS) is not well studied in BF-WSNs. Existing works always have a rigid assumption that there is only one single query which is targeted at the whole network. Aiming at making the work more practical and general, we investigate the general MLAS problem in BF-WSNs, which is targeted at any subset of nodes in the network and aimed for an arbitrary number of aggregation queries. First, the general MLAS problem when there is one single query is studied. To control the number of nodes participating in the aggregation process, a node selection algorithm is proposed to cover and connect the whole target nodes. Then, a latency and energy aware scheduling algorithm is proposed to integrate the construction of aggregation tree with the chosen nodes, and the computation of a conflict-free schedule simultaneously, relying on non-predetermined structures. Second, the general MLAS problem when there is a group of aggregation queries is studied. Through designing some special structures to avoid collisions between both current and existing aggregation schedules, an algorithm without any waiting time is proposed. Additionally, the algorithm under physical interference model and dynamic energy arrival model are also presented. The theoretical analysis and simulation results verify that the proposed algorithms have high performance in terms of latency and energy efficiency.

Broadcasting is an essential operation in wireless networks for disseminating the message from the source node to all other nodes. Unfortunately, the problem of Minimum Latency Broadcast Scheduling (MLBS) in duty-cycled wireless sensor networks is not well studied. In existing works, the construction of broadcast tree and the scheduling of transmissions are conducted separately, where a tree-based structure is used as the input of the scheduling algorithm. Relying on a pre-determined tree may result in a much large latency even using the optimal scheduling method. Thus, the MLBS problem in duty-cycled WSNs without the above limitation is investigated in this paper. First, to avoid relying on a pre-determined structure, a two-step scheduling algorithm is proposed to construct the broadcast tree and compute a collision-free schedule simultaneously. To the best of our knowledge, this is the first work that can integrate these two kinds of operations together. Second, a novel transmission mode, i.e., concurrent broadcasting, is first introduced for wireless networks and several techniques are designed to further improve the broadcast latency. Third, the multiple messages broadcasting and all-to-all broadcasting algorithms, which can generate a series of broadcast schedules independently without a pre-determined tree, are also proposed by taking care of the collisions in both the current and the previous broadcast schedules. Finally, the theoretical analysis and experimental results demonstrate the efficiency of the proposed algorithms in terms of latency.

Graph-based methods have achieved impressive performance on semisupervised classification (SSC). Traditional graph-based methods have two main drawbacks. First, the graph is predefined before training a classifier, which does not leverage the interactions between the classifier training and similarity matrix learning. Second, when handling high-dimensional data with noisy or redundant features, the graph constructed in the original input space is actually unsuitable and may lead to poor performance. In this article, we propose an SSC method with novel graph construction (SSC-NGC), in which the similarity matrix is optimized in both label space and an additional subspace to get a better and more robust result than in original data space. Furthermore, to obtain a high-quality subspace, we learn the projection matrix of the additional subspace by preserving the local and global structure of the data. Finally, we intergrade the classifier training, the graph construction, and the subspace learning into a unified framework. With this framework, the classifier parameters, similarity matrix, and projection matrix of subspace are adaptively learned in an iterative scheme to obtain an optimal joint result. We conduct extensive comparative experiments against state-of-the-art methods over multiple real-world data sets. Experimental results demonstrate the superiority of the proposed method over other state-of-the-art algorithms.

As computing systems continue to increase in complexity, energy optimization plays a key role in the design and implementation of heterogeneous systems. Although the energy consumed by off-chip memory accounts for a large proportion of the total power consumed by the system as a whole, current research on energy optimization mainly focuses on optimizing the energy consumed by the processors. This article explores the coordinated optimization of the holistic performance of the processors and memory system for heterogeneous systems with energy constraints. A communication–computing pipeline model for parallel executions is characterized to optimize program performance by simultaneously scaling the voltage and frequency of the processors and memory using task allocation strategies. A synergistic load-balancing optimization approach is presented to resolve the load imbalance among graphics processing units. Our experimental results substantiate the effectiveness of the approach in terms of execution times and throughputs with the energy constraints.

Data aggregation is a fundamental yet popular operation in wireless networks where the sink needs to obtain the combined information of the whole network. However, the problem of minimum latency aggregation scheduling (MLAS) is not well studied in cognitive radio networks. Few studies have addressed this issue and most previous aggregation methods all assume that a fixed-structure based aggregation tree is constructed in advance, which may result in the selection of a node with limited spectrum opportunities as the parent by many nodes and by extension results in a large latency. Thus, the MLAS problem in cognitive radio networks (MLAS-CR) without the above limitation is investigated in this paper. First, the MLAS-CR problem with primary social behaviors where the activity of primary users can be predicted is studied. To make full use of the limited spectrum opportunities, we integrate the construction of the aggregation tree, and the computation of a conflict-free schedule simultaneously, without any predetermined structures. Second, the MLAS-CR problem without the above assumption is also investigated. To reduce the latency, a two-way aggregation scheduling method is proposed to adaptively choose the parent with only current channel information. To further reduce the latency, we also introduce a new data aggregation mode for CRN, i.e., Data Aggregation Scheduling in The Dark, to utilize the spectrum opportunities of scheduled nodes. Finally, the theoretical analysis and simulation results verify that the proposed algorithms have high performance in terms of latency.

The emerging energy-harvesting technology enables charging sensor batteries with renewable energy sources, which has been effectively integrated into Wireless Sensor Networks (EH-WSNs). Due to the limited energy-harvesting capacities of tiny sensors, the captured energy remains scarce and differs greatly among nodes, which makes the data aggregation scheduling problem more challenging than that in energy-abundant WSNs. In this article, we investigate the Minimum Latency Aggregation Scheduling (MLAS) problem in EH-WSNs. First, we identify a new kind of collision in EH-WSNs, named as energy-collision, and design several special structures to avoid it during data aggregation. To reduce the latency, we try to choose the parent adaptively according to nodes’ transmission tasks and energy-harvesting ability, under the consideration of collisions avoidance. By considering transmitting time, residual energy, and energy-collision, three scheduling algorithms are proposed under protocol interference model. Under physical interference model, several approximate algorithms are also designed by taking account of the interference from the nodes several hops away. Finally, the theoretical analysis and simulation results verify that the proposed algorithms have high performance in terms of latency.

Broadcasting is a fundamental networking service where the source node tries to disseminate the message to the whole network. The problem of Minimum Latency Broadcast Scheduling (MLBS) which seeks a fast and collision-free broad-cast schedule has been well studied when nodes are energy-abundant. However, in battery-free wireless networks, node can only receive or transmit packets after it has harvested enough energy. In such networks, it is of great importance to exploit the harvested energy smartly to reduce broadcast latency. Un-fortunately, the existing works rely on predetermined structures may greatly increase the latency by choosing a node with large charging latency as the backbone node. In addition, they assume each node can only transmit once which may result in much waiting latency. To address the above issues, we investigate the MLBS problem in battery-free wireless networks without predetermined structures in this paper. Firstly, to make use of the harvested energy smartly, we intertwine the construction of broadcast tree and the computation of an energy-satisfied and collision-free schedule. Secondly, a Delayed Broadcasting technique is proposed for each node to tradeoff between the number of transmissions and its waiting latency. By considering residual energy and transmitting time, two latency and energy aware scheduling algorithms are proposed, in which the broadcast tree can be constructed adaptively according to nodes' energy status. Finally, the theoretical analysis and simulation results verify that the proposed algorithms have high performance in terms of broadcast latency.

Eliminating haze interference in images is still a challenging problem. In this paper, we consider more systematically the physical hazing mechanisms, combined with deep learning, propose a new end-to-end dehazing network called DHD-Net. For physical hazing mechanisms, we fuse the global atmosphere light, transmission maps, and the atmospheric scattering model for dehazing. For the estimation of global atmosphere light, We propose a deep learning-based haze density estimation algorithm (DL-HDE). We establish a new dataset, of which each data item consists of the hazy image, the transmission map, the haze-free image, and the dense-haze area mask. Our experimental results demonstrate that our proposed DHD-Net has better dehazing performance than state-of-the-art algorithms.

Beaconing is a fundamental networking service where each node broadcasts a packet to all its neighbors locally. Unfortunately, the problem Minimum Latency Beaconing Schedule (MLBS) in duty-cycled scenarios is not well studied. Existing works always have rigid assumption that each node is only active once per working cycle. Aiming at making the work more practical and general, MLBS problem in duty-cycled network where each node is allowed to active multiple times in each working cycle (MLBSDCA for short) is investigated in this paper. First, a novel kind of coloring problem, named as label coloring problem, is identified and analyzed. Second, an edge-based scheduling framework is designed and the MLBSDCA under protocol interference model is transformed to such coloring problem. Based on label coloring, a group first-fit scheduling algorithm is designed for MLBSDCA under protocol interference model. After that, a (ρ + 1)2|W|-approximation algorithm is proposed to further reduce the beaconing latency, where p denotes the interference radius, and |W| is the maximum number of active time slots per working cycle. When p and |W| is equal to 1, the approximation ratio is only 4, which is better than the one (i.e., 10) in existing works. Furthermore, two approximation algorithms for MLBSDCA under physical interference model are also investigated. The theoretical analysis and experimental results demonstrate the efficiency of the proposed algorithms in term of latency.

Broadcasting is a fundamental networking service where the source node disseminates the message to all the other nodes. Unfortunately, the problem of Minimum Latency Broadcast Scheduling (MLBS) in duty-cycled wireless networks is not well studied. In the existing works, the construction of broadcast tree and the scheduling of transmissions are conducted separately, which may result in a bad-structured broadcast tree and then a large latency is obtained even using the optimal scheduling method. Thus, the MLBS problem in duty-cycled wireless networks without above limitation is investigated in this paper. Firstly, a Two-Step Scheduling algorithm is proposed to construct the broadcast tree and compute a collision-free schedule simultaneously. The proposed method can generate a latency-aware broadcast tree adaptively to reduce the broadcast latency. To the best of our knowledge, this is the first work that can integrate these two kinds of operations together. Additionally, a novel transmission mode, i.e., concurrent broadcasting, is first introduced in wireless networks and several techniques are designed to further improve the broadcast latency. Finally, the theoretical analysis and experimental results demonstrate the efficiency of the proposed algorithms in term of latency.

Broadcasting is a fundamental function for disseminating messages in multihop wireless networks. Minimum-Transmission Broadcasting (MTB) problem aims to find a broadcast schedule with minimum number of transmissions. Previous works on MTB in duty-cycled networks exploit a rigid assumption that nodes have only active time slot per working cycle. In this paper, we investigated the MTB problem in duty-cycled networks where nodes are allowed arbitrary active time slots per working cycle (MTBDCA problem). Firstly, it is proved to be NP-hard and o(ln⁡Δ) -inapproximable, where Δ is the maximum degree in the network. Secondly, an auxiliary graph is proposed to integrate nodes’ active time slots into the network and a novel covering problem is proposed to exploit nodes’ multiple active time slots for scheduling. Then, a ln⁡(Δ+1) -approximation algorithm is proposed for MTBDCA and a (ln⁡(Δ+1)+Δ) -approximation algorithm is proposed for all-to-all MTBDCA. Finally, extensive experimental results demonstrate the efficiency of the proposed algorithm.

The abundance of a mixed pixel of certain class can be understood as to get the relative score referring to the pure representative of this class, while not be classified with two absolute and discrete value as ”lor 0”. This is in accordance with the Relative Attribute Learning (RAL) problem in computer vision. In RAL, the concept of “relative attribute” is used to describe the belonging level of an obj ect to certain class with a score which is achieved from a learn-to-rank problem using rankSVM framework. To utilize information between data samples and even of mixed pixels, Relative Attribute based Unmixing (RAU) is proposed first time by using relative attribute to describe the abundance of mixed pixel as relative purity of certain class and learn the abundance with rankSVM. The mixed data sample are used to construct training comparisons set in rankSVM with archetypes generated by the reported Kernel Archetypal Analysis (KAA) unmixing method. In addition, spectral variability is also addressed by constructing comparisons set with synonyms spectrum achieved from KAA. Experiments on both synthetic and real hyperspectral mixed image have demonstrated the potential value of proposed method for mixed pixel analysis.

The emerging energy harvesting technology enables charging sensor batteries with renewable energy sources, which has been effectively integrated into Wireless Sensor Networks (EH-WSNs). Meanwhile, data aggregation is an essential operation in a WSN. The problem of Minimum Latency Aggregation Scheduling (MLAS) which seeks a fast and collision-free aggregation schedule has been well studied when nodes are energy-abundant. However, due to the limited energy harvesting capacities of tiny sensors, the captured energy remains scarce and differs greatly among nodes. Thus, all of the previous algorithms for MLAS are not suitable in EH-WSNs. In this paper, we investigate the MLAS problem in EH-WSNs. To make use of the harvested energy smartly, we construct an aggregation tree adaptively according to the residual battery level at each node. Furthermore, we identify a new kind of collision, named as energy -collision, and design a special structure to assist in avoiding it. By considering transmitting time, residual energy, and energy-collision, we propose three scheduling algorithms for MLAS problem in EH-WSNs. The theoretical analysis and simulation results verify that the proposed algorithms have high performance in terms of aggregation latency compared with the baseline methods.

In this paper, we propose a novel concordance coefficient, called order statistics concordance coefficient (OSCOC), to quantify the association among multichannel biosignals. To uncover its properties, we compare OSCOC with three other similar indexes, i.e., average Pearson's product moment correlation coefficient (APPMCC), Kendall's concordance coefficients (KCC), and average Kendall's tau (AKT), under a multivariate normal model (MNM), linear model (LM), and nonlinear model. To further demonstrate its usefulness, we present an example on atrial arrhythmia analysis based on real-world multichannel cardiac signals. Theoretical derivations as well as numerical results suggest that 1) under MNM and LM, OSCOC performs equally well with APPMCC, and outperforms the other two methods, 2) in nonlinear case, OSCOC even has better performance than KCC and AKT, which are well known to be robust under increasing nonlinear transformations, and 3) OSCOC performs the best in the case study of arrhythmia analysis in terms of the volume under the surface.

Summary With the continued development of hardware and software, graphics processing unit (GPU) has been used in general purpose computational fields, while accelerating applications for CPUs. To achieve high computing performance, a GPU typically includes hundreds of computing units. The high density of computing resource on‐chip incurs high power consumption as well as engendering high performance. The power consumption problem has become one of the most important problems for the development of GPUs. Focusing on a CPU‐GPU heterogeneous parallel system, this research proposed an architecture‐level GPU energy model, with the aim of reducing system energy demand and improving system efficiency. Taking the influences of memory and temperature on GPU energy demand into account, a dynamic energy model based on division of computation and memory and a static energy model based on real‐time temperature perception were established. Validation, through the evaluation and comparative analysis of nine typical GPU programmes, demonstrated that these models could reduce chip energy demand under the performance constraint conditions imposed. Copyright © 2014 John Wiley & Sons, Ltd.

The inadequate resource allocation, lack of I/O performance prediction and insufficient isolation are affecting the storage performance in the multi-tenant cloud storage environment. In order to guarantee the Quality of Service (QoS), Softwaredefined Storage (SDS) is an effective approach in data centers. However, the lack of intelligence, robustness and selfadjustment are blocking the applications and promotions of SDS heavily. This paper focuses on the QoS-Aware I/O resource scheduling problem to build data centers with high availability, scalability and QoS. We will study workload characteristics, requirement analysis, the theory of QoS in SDS and I/O scheduling strategies. We obtain such goals by proposing a mathematics model of workload burstness, QoS semantic description with rule execution mechanisms and dynamic robust I/O scheduling algorithms for multi-type resources allocation. In the current progress, A QoS-Aware I/O Scheduling Framework towards SDS, qSDS has been proposed for the SSD/HDD hybrid storage. The preliminary evaluation in some benchmarks shows that qSDS can gain better performance compared with other strategies.