Search

In this letter, we propose a novel dynamic fusion model (DFM) with joint variational and deep priors for the task of hyperspectral image super-resolution (HISR). The given model can benefit from both the advantages of traditional modeling and deep learning (DL) methods, thus achieving significant improvements based on existing deep pretrained models. Specifically, the given model mainly contains two newly designed terms, i.e., the weighted spatial fidelity (WSF) term and the deep fusion (DF) term. The WSF term focuses on the spatial recovery of the low-resolution hyperspectral image (LR-HSI) through the high-resolution multispectral image (HR-MSI) without the knowledge of the spectral response matrix, thus the proposed DFM can be viewed as a semi-blind model for HISR. Moreover, the DF term relied upon DF with a designed adaptive weight matrix, which can effectively inject the deep priors into the traditional minimization model. Besides, the proposed DFM can be quickly and effectively solved using the alternating direction method of multipliers (ADMM). Experimental results on widely used datasets demonstrate the superiority of our approach compared with state-of-the-art HISR methods. [ABSTRACT FROM AUTHOR]

Microarc oxidation (MAO) is a popular surface treatment process to generate oxide coatings with excellent mechanical properties on valve metals. As a plasma discharge technique, the discharge mechanism in MAO is different from that in the bipolar plate. Due to the alternating arcs and multiple electrolytes in MAO, it is difficult to control and optimize the coating properties. Based on the arcing mechanism and ion motion, the boundary conditions of no-arc discharge, alternating arc discharge, and continuous arc discharge are derived, and the relationship between the discharge current and breakdown current and sustained arc current is investigated. For the same electrolyte, the boundary conditions are determined by equivalent concentration and conductivity. The results show that in the stable alternating arc discharge regime, the higher the concentration and smaller the conductivity, the more intense is the discharge. According to the boundary conditions, MAO experiments are designed using phosphate electrolytes to improve the hardness of oxide coatings. By selecting electrolyte characteristics close to the continuous arc discharge boundary in the stable alternating arc discharge regime, the arc discharge is most intense and brightest and the largest energy is transferred to the Al2O3 coating. Consequently, the crystallinity and mechanical properties are improved significantly. [ABSTRACT FROM AUTHOR]

In permanent-magnet synchronous motor (PMSM) drives, many current control strategies are presented to compensate the effects of rotor flux distortion on torque performance. These strategies are usually realized in the synchronously rotating reference frame and at least two-phase current sensors are used. Although the technique of phase current reconstruction enables three-phase currents to be obtained with a single dc-link current sensor, two-phase current controllers and complex coordinate transformations are still necessary. In this article, a simplified current control strategy for low-power PMSM drives is proposed. First, a torque allocation law is designed to resolve the reference electromagnetic torque into the reference three-phase currents with compensation for rotor flux distortion. Second, the reference three-phase currents are further converted into the reference dc-link current according to the averaging theory. Third, the amplitude and waveform control of three-phase currents are fulfilled by using dc-link current controller and ac-side pulsewidth modulation, respectively. This strategy can compensate the effects of rotor flux distortion on torque performance and is easy to implement with only one current sensor and one controller. Simulation and experimental results demonstrate the effectiveness of the proposed strategy. [ABSTRACT FROM AUTHOR]

Pansharpening refers to a spatio-spectral fusion of a lower spatial resolution multispectral (MS) image with a high spatial resolution panchromatic image, aiming at obtaining an image with a corresponding high resolution both in the domains. In this article, we propose a generic fusion framework that is able to weightedly combine variational optimization (VO) with deep learning (DL) for the task of pansharpening, where these crucial weights directly determining the relative contribution of DL to each pixel are estimated adaptively. This framework can benefit from both VO and DL approaches, e.g., the good modeling explanation and data generalization of a VO approach with the high accuracy of a DL technique thanks to massive data training. The proposed method can be divided into three parts: 1) for the VO modeling, a general details injection term inspired by the classical multiresolution analysis is proposed as a spatial fidelity term and a spectral fidelity employing the MS sensor’s modulation transfer functions is also incorporated; 2) for the DL injection, a weighted regularization term is designed to introduce deep learning into the variational model; and 3) the final convex optimization problem is efficiently solved by the designed alternating direction method of multipliers. Extensive experiments both at reduced and full-resolution demonstrate that the proposed method outperforms recent state-of-the-art pansharpening methods, especially showing a higher accuracy and a significant generalization ability. [ABSTRACT FROM AUTHOR]

This article proposes a cognitive radar (CR) framework to improve the performance of the estimation and recognition for range-extended targets. To overcome the angular uncertainty and exploit the temporal correlation of target signatures, we introduce the static multimodel approach and the exponential correlation model to describe the dynamics of target signatures. The estimate of target signatures and the probability distribution of target hypotheses are updated recursively in response to the latest echo, ruled by the Bayesian theory. A waveform optimization module is integrated into the framework, which designs the waveform for the next emission according to the knowledge extracted from the echo history. A design criterion based on mutual information (MI) is derived, which implies the average estimation capability for all target hypotheses and the mean between-hypotheses divergence. The proposed method is examined using both synthetic data and electromagnetic (EM) simulation data of realistic targets. [ABSTRACT FROM AUTHOR]

Jitter is one of the crucial factors used to characterize high-speed serial links and integrated circuit performance. Jitter decomposition is key tool with which to characterize jitter at a given bit error rate (BER). In this article, a novel jitter decomposition algorithm is proposed using convolutional neural networks to decompose random jitter (RJ) and deterministic jitter (DJ) by images of a jitter histogram, and predict the total jitter (TJ) at a BER of 10−12. The jitter histogram – that is, training data – is obtained by modeling the high-speed serial link in an advanced design system. The test results verify the feasibility and accuracy of the proposed method. The RJ mean absolute error (MAE) of the test is 0.8721 ps and the error rate 6.74%, and the DJ MAE is 4.8684 ps and the error rate 2.03%, and TJ MAE is 8.9406 ps and the average error rate 2.13%. In addition, results show that the effectiveness of the proposed method is better than most other jitter decomposition methods by evaluating the average error rate. [ABSTRACT FROM AUTHOR]

Phase-locked loop (PLL) based resolver-to-digital conversion (RDC) can provide high-precision information of angular position and velocity for servo control system with resolver. However, there exist considerable harmonic components and random noises in the sinusoidal pair signals from resolver, which may impede further enhancement on the precision of RDC. In this paper, a signal filtering method is presented for RDC by combining two second-order generalized integrators (SOGIs) with a frequency-locked loop (FLL). In this method, FLL is designed in the form of second order to obtain high-accuracy estimate of resolver signal frequency, and two SOGIs are employed as frequency-adaptive band-pass filters to filter resolver pair signals by using the estimated frequency as the center frequency. Hence, the proposed method can track the fundamental components of resolver pair signals accurately, and thus harmonic components and noises can be reduced. Compared with the conventional SOGI-FLL, this method has better frequency adaptation and higher tracking accuracy since the accuracy of frequency estimation is improved by using second-order FLL. It resolves the problem of PLL in suppressing harmonics and low-frequency noises, thereby improving the accuracy of RDC. Simulation and experimental results demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

A compact high-speed active quenching and recharging circuit (AQRC) is proposed for single-photon avalanche diode (SPAD) detectors fabricated in standard 0.18 μm CMOS technology. A novel sensing and quenching scheme is employed to speed up the process of avalanche detection, enabling a low afterpulsing probability and a high photon-counting rate. The simulation and experimental results indicate that the sensing and quenching time is reduced down to 0.7 ns and the maximum photon-counting rate is close to 200 Mcps (counts per second), meanwhile the afterpulsing probability as low as 0.75% is obtained. Due to the compact configuration, the area occupation of the AQRC is only 306 μm2. Such outstanding advantages suggest that the proposed AQRC is very suitable for the miniaturized and high-speed SPAD-array detectors. [ABSTRACT FROM AUTHOR]

With the development of Internet of Things (IoT) technology, people’s demand for indoor location-based services is growing. Indoor magnetic field has a wealth of fingerprint information that can be combined with position stamps to construct an indoor magnetic map (IMM). Localization scenarios based on IMM include personnel attendance, equipment tracking, and fire emergency, etc. However, the localization technology is complicated, the localization accuracy based on IMM is generally low, and the localization result lacks credibility due to the fluctuation of indoor magnetic field. Therefore, it is urgent to study a novel localization method in terms of the tradeoff among complexity, accuracy and credibility, which is the purpose of this article. First, the Euclidean distance constraint (EDC) with a variable search radius is proposed to make a rough estimate of the entity’s location. Then the iterative interpolation method (IIM) is proposed to refine the local-IMM, and the multi-magnetic- fingerprint-fusion (MMFF) is proposed to match the magnetic fingerprint based on the refined local-IMM. In short, the proposed method is called focusing matching localization method by combining coarse estimation with fine estimation, which can be abbreviated as EDC-IIM-MMFF. Finally, the effectiveness of the proposed localization method is verified by a series of experiments. The tests show that the root mean square error (RMSE) of location solved by the proposed method is less than 0.3 meters, and the 99% cumulative distribution function (CDF) of location error is less than 0.5 meters, and the complexity is lower than the existing methods, such as Gaussian process regression (GPR), extended Kalman filter (EKF), particle filter (PF), etc. [ABSTRACT FROM AUTHOR]

Active sensing systems prefer probing waveforms with good auto-correlation properties. Faced with the difficulty in getting individual sequences with impulse-like aperiodic auto-correlation functions, we resort to the temporal waveform diversity based on complementary sets of sequences (CSS). It seems an ideal solution, but has rarely been used due to the extreme sensitivity of CSS to Doppler shifts. This paper is devoted to applying numerical methods to the design of Doppler resilient complementary waveforms (DRCW) whose ambiguity functions are almost free of range sidelobes along modest Doppler shifts. We formulate the problem as minimizing the worst-case peak sidelobe level (PSL) with low peak-to-average power ratio (PAR) constraints, and then approximate the min-max problem by a ${l}_{p}$ -norm minimization problem. A hierarchical strategy is developed to tackle the phase optimization and the amplitude-phase optimization respectively. Correspondingly, two algorithms based on phase-gradient and Majorization-Minimization are proposed, both of which can be efficiently computed via FFT. Numerical experiments show our method is effective and superior to existing ones. The DRCWs thus obtained have their range sidelobes at all lags suppressed below −60dB, and are expected to bring significant performance improvements in active sensing. Two application examples, detecting weak targets in heavy clutter and estimating the high-resolution range profile of an extended target by active radars, are given. [ABSTRACT FROM AUTHOR]

This article proposes a fuzzy robust weighted method (FRWM) for a bilevel game model comprising both multiple leaders and followers (named multi-criteria bilevel game). In this game, each decision maker has several competing objectives and plays a noncooperative multi-criteria Nash game with others at his level. Each decision maker is assumed to be uncertain about the exact weights over his objectives, but these weights belong to a given set. Furthermore, each player proposes an FRWM to address the uncertainty, i.e., each player will minimize his maximum weighted sum objective in which the maximization is in regards to the given weight set. Then, a new equilibrium concept called a fuzzy robust weighted Nash equilibrium (FRWNE) is presented. It can be proven at least one this equilibrium is present even though the weights are infinite. When the weight set of multi-criteria bilevel game is polyhedral, we can obtain an FRWNE by solving a group of mathematical programming problems with equilibrium constraints. We illustrate the usefulness and efficiency of our fuzzy robust weighted approach to a supply chain multi-criteria bilevel competition problem. [ABSTRACT FROM AUTHOR]

In software-based resolver-to-digital conversion, it is necessary to detect the sinusoidal and cosinusoidal envelopes from the amplitude-modulated resolver signals in advance. The synchronous peak sampling method is usually used. However, the sampling frequency is limited by the excitation frequency in this method, which may lead to large quantization errors in the detected envelopes. To solve this problem, an oversampling synchronous envelope detection method is proposed in this article. First, the resolver signals are sampled at multiple fixed phase angles within each excitation period by synchronous oversampling. Second, the oversampled resolver signals are divided into multiple envelope pairs for normalization and correction. Finally, the calibrated envelope pairs are integrated into a complete pair of sinusoidal and cosinusoidal envelopes of the angular position. Thus, the envelope detection of resolver signals is implemented. Compared with the synchronous peak sampling, the proposed method can effectively reduce the quantization errors of the detected envelopes and the measurement accuracy of the rotor angular position can be further improved. Simulation and experimental results demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]