Network embedding aims to preserve vertex similarity in an space. Existing approaches usually define the by direct links or common neighborhoods between nodes, i.e. structural equivalence. However, vertexes which reside different parts of network may have similar roles positions, regular equivalence, is largely ignored literature embedding. Regular equivalence defined a recursive way that two regularly equivalent neighbors are also equivalent. Accordingly, we propose new approach named Deep...

Network embedding has recently attracted lots of attentions in data mining. Existing network methods mainly focus on networks with pairwise relationships. In real world, however, the relationships among points could go beyond pairwise, i.e., three or more objects are involved each relationship represented by a hyperedge, thus forming hyper-networks. These hyper-networks pose great challenges to existing when hyperedges indecomposable, that is say, any subset nodes hyperedge cannot form...

This paper presents an energy-efficient and terrain-information-and-preceding-vehicle-information-incorporated energy management strategy for fully electric vehicles (FEVs) equipped with in-wheel motors. Saving driving terrain preview preceding vehicle movement prediction are crucial to prolong the distance FEV. Unlike conducting optimization under assumption that movements already known in most studies, front predicted during each control cycle based on vehicle-to-vehicle communication, FEV...

The ability to perform long-range pedestrian detection is essential for autonomous vehicles. However, 3-D LIDAR, an object's point cloud becomes sparse when it away, directly affecting its as a result. In this paper, novel density enhancement method proposed improve the quality of cloud. input raw cloud; first, high-quality local coordinate system built using new evaluation metric, and then radial basis function (RBF)-based interpolation performed based on system. Finally, resampling...

In the post-deep learning era, Transformer architecture has demonstrated its powerful performance across pre-trained big models and various downstream tasks. However, enormous computational demands of this have deterred many researchers. To further reduce complexity attention models, numerous efforts been made to design more efficient methods. Among them, State Space Model (SSM), as a possible replacement for self-attention based model, drawn in recent years. paper, we give first...

Subspace clustering aims to partition the data points drawn from a union of subspaces according their underlying subspaces. For accurate semisupervised subspace clustering, all that have must-link constraint or same label should be grouped into subspace. However, this is not guaranteed in existing approaches. Moreover, these approaches require additional parameters for incorporating supervision information. In paper, we propose constrained low-rank representation (CLRR) robust based on novel...

This paper presents an improved correlation-based timing-skew calibration strategy with constant input impedance characteristics. A full sampling rate operating time-interleaved reference ADC (TI-RADC) whose interleaving factor is coprime TI-ADC introduced to replace the conventional single-channel ADC. theoretically demonstrates that by setting appropriate time interval between edge of and TI-RADC aligning each channel in other but not ADC, skew would affect timing TI-ADC. prototype 13-bit...

This brief presents a two-channel 14-bit 500-MS/s time-interleaved analog-to-digital converter (ADC) with background time skew calibration fabricated in 0.18-μm CMOS process. Time is detected an improved autocorrelation-based strategy and corrected digitally controlled delay line. Different from traditional strategy, the proposed doesn't align all channel ADCs reference ADC, but aligns each other remains interval ADC. The acceptable wide range. It can enhance sensitivity of detection. By...

Abstract ZY 3‐02 is the first Chinese earth‐observation satellite equipped with a laser altimeter, designed to assist stereo‐image mapping elevations. Launch vibrations, space environment and other factors, both expected unexpected, cause bias in ranging pointing of influencing its height measurement accuracy. This paper proposes an in‐orbit geometric calibration model which considers this as unknown systematic errors. The parameters were calibrated by minimum error principle using spot...

Abstract: The constrained ordination method from quantitative ecology was utilized to assess the relationship between landscape patterns and nonpoint‐source (NPS) pollution for purpose of identifying effective water‐quality improvement practices in Danjiangkou Reservoir (DJKR) basin, China. soil water assessment tool (SWAT) applied simulate NPS Fragstats model calculate metrics. study concluded that organic nutrients formed main pollutant DJKR basin most occurred along with loss. Based on...

This article presents a 14-bit 200-Ms/s pipelined analog-to-digital converter (ADC) for wide input frequency range. The ADC adopts sample-and-hold amplifier-less (SHA-less) 3.5-bit front-end stage. A dedicated path combining architecture is proposed the flash sub-ADC-based pipeline to reduce aperture error, and expand up 490 MHz under sampling rate. To further improve accuracy power, capacitor splitting comparator interpolation techniques are used in first has been designed fabricated 180-nm...

Privacy leakage of Internet Things (IoT) has become a great challenge with the popularity IoT services through mobile networks, such as smart homes, wearables, and healthcare. While previous work summarized general structures to analyze privacy provide case studies specific devices or scenarios, it is still challenging conduct comprehensive systematic quantification study large-scale in real world. To combine analyses real-world measurements, we method quantify on network traffic data set...

The functional generalized additive model (FGAM) provides a more flexible nonlinear regression than the well-studied linear model. This paper restricts attention to FGAM with identity link and errors, which we will call model, generalization of studies minimax rate convergence predictions from in framework reproducing kernel Hilbert space. It is shown that optimal determined by decay eigenvalues specific function, turn joint distribution any two points random predictor function. For special...

This brief presents a simple capacitor mismatch calibration in SAR ADCs, which is foreground and on chip. Capacitor errors are extracted based output histogram under triangular input signal generated chip compensated by tuning capacitors with DAC. A novel one-by-one strategy can simplify the extraction isolating individual contribution of each error. The proposed needs no computation validated 10-bit 100MS/s ADC. Size ADC core including system only 90μm×97μm. Measurement results show...

Objectives To analyze the epidemiological characteristics and etiology of crayfish-related rhabdomyolysis. Methods Cases rhabdomyolysis in Wuhan were monitored, professional training city’s surveillance personnel was conducted. Unified questionnaires used to collect data. Results The first case occurred on July 12, 2016. Subsequently, 423 patients reported over next 7 years, with muscle pain, weakness, chest distress as main symptoms. In total, 64.54% (273/423) females, young adults (aged...

Nonnegative matrix factorization (NMF) as a popular technique to find parts-based representations of nonnegative data has been widely used in real-world applications. Often the which these applications process, such motion sequences and video clips, are with ordered structure, i.e., consecutive neighbouring samples very likely share similar features unless sudden change occurs. Therefore, traditional NMF assumes be independently distributed, making it not proper for analysis data. In this...

Trust prediction, aiming to predict the trust relations between users in a social network, is key helping discover reliable information. Many prediction methods are proposed based on low-rank assumption of network. However, one typical property network that follow power-law distribution, i.e., few trusted by many other users, while most tail have trustors. Due these fundamental made existing seriously violated and becomes unrealistic. In this paper, we propose simple yet effective method...

The modeling of scalar mixing timescale remains a primary challenge in the transported probability density function (TPDF) method. variation among species, i.e., differential mixing, results from difference molecular diffusivity and reaction-induced gradient. Nevertheless, vast majority TPDF studies on turbulent non-premixed flames simply apply single determined by mixture fraction. In this work, (RIDM) model for individual species is proposed. key idea RIDM to approximate relative magnitude...

As high-dynamics and weak-signal are of two primary concerns navigation using Global Navigation Satellite System (GNSS) signals, an acquisition algorithm based on three-time fractional Fourier transform (FRFT) is presented to simplify the calculation effectively. Firstly, correlation results similar linear frequency modulated (LFM) signals derived basis high dynamic GNSS signal model. Then, principle obtaining optimum rotation angle analyzed, which measured by FRFT projection lengths with...

Nonnegative matrix factorization (NMF), which aims at finding parts-based representations of nonnegative data, has been widely applied to a range applications such as data clustering, pattern recognition and computer vision. Real-world are often sparse noisy may reduce the accuracy representations. And small portion have prior label information, which, if utilized, can improve discriminability In this paper, we propose robust semi-supervised (RSSN-MF) approach takes all factors above into...

A 14-bit 250MS/s low power pipeline analog-to-digital converter (ADC) implemented in a 0.18μιη CMOS process is presented this paper. SHA-less 3.5-bit front-end adopted to achieve design. An aperture error eliminating technique and flash ADC optimization design techniques such as capacitor splitting, interpolation, offset calibration, are both used wideband input even for the with resolution up 3.5-bit. The post simulation results show achieves an SNR of 74.6 dB, SNDR 74.4 dB SFDR 87.1 70MHz...