Search

Heterogeneous SoCs integrate FPGAs and microprocessor cores on the same fabric to accelerate applications, such as cryptography and deep learning. Since FPGAs share resources with the microprocessor cores, they can launch noncacheable synchronous DRAM (SDRAM) transactions through direct FPGA-to-microprocessor SDRAM interface. Therefore, if the FPGA 3rd party IPs (3PIPs) are malicious, they can launch rowhammer attacks on the SDRAM. Today’s countermeasures based on performance counters cannot detect these attacks because memory transactions from FPGAs do not pass through the cache. In addition, today’s countermeasures that count the frequency of activation of memory rows cannot identify the intellectual property (IP) that launches the attack from the FPGA. We present a security solution that monitors the SDRAM transactions from IPs on the FPGA to each bank of the microprocessor SDRAM through the FPGA-to-microprocessor SDRAM interface. The proposed monitor is implemented on the FPGA fabric. It can detect attempts to launch a rowhammer attack before it causes bit flips in the SDRAM. It utilizes 6.3% of the adaptive logic modules (ALMs) available in an Intel Cyclone V FPGA, when multiple IPs are monitored. [ABSTRACT FROM AUTHOR]

Imaging of a subwavelength microstructure made of a periodic grid-like finite set of circular rods is carried out from transient scattered field data in different configurations of sources and receivers. The goal is to identify the position of possibly missing rods. Time reversal is confirmed as a cheap yet efficient first-order diagnostic method even in the demanding context of a subwavelength microstructure. Tools of deep learning, expected to be valid in general circumstances if data acquired in sufficient numbers, are in parallel developed to image the microstructure. To that effect, recurrent neural networks (RNN) and convolutional neural networks (CNN) are both used. Pros and cons of all approaches are illustrated by comprehensive simulations from synthetic data computed via a finite difference time domain method (FD-TD) software carefully tailored to the microstructure model and used also to make the networks learn the microstructures. The analysis is completed from examples of laboratory-controlled data acquired on a microstructure prototype set within a microwave anechoic chamber. These examples confirm the good promises of neural networks even with rather scarce data as exemplified in a forward scattering case—fixed source and receiver antennas face each other and the microstructure is rotated between them. [ABSTRACT FROM AUTHOR]

The oscillator fluctuations are described as the phase or frequency-noise spectrum or in terms of a wavelet variance as a function of the measurement time. The spectrum is generally approximated with the “power law,” i.e., a Laurent polynomial with integer exponents of the frequency. This article provides: 1) the analytical expression of the response of the wavelet variance parabolic variance (PVAR) to the frequency-noise spectrum in the general case of noninteger exponents of the frequency and 2) a useful approximate expression of the statistical uncertainty. In turn, PVAR is relevant in that it replaces the widely used modified Allan variance (MVAR), featuring the identification of noise processes with fewer data. [ABSTRACT FROM AUTHOR]

In this work, a fully physics-based stochastic 3-D modeling of metal–oxide-based resistive random access memory (RRAM) device has been conducted. In contrast to general 1-D or 2-D modeling of RRAM devices, 3-D modeling is able to provide a picture of the evolution of the 3-D random distributions of oxygen vacancies and temperature inside the device during the forming and SET/RESET processes. The modeling confirms that a tunneling gap between the top electrode (TE) which serves as an oxygen ion reservoir and the tip of the conductive filament may form at a high resistive state, which is assumed in the frequently used compact models of RRAM devices. The modeling shows that the sites with high temperature are concentrated in a cylindrical region at the center of the RRAM device, running from the inert bottom electrode to the TE. The modeling can reproduce both typical current-voltage characteristics of metal–oxide RRAM devices in forming and RESET and SET operations under a dc voltage sweep and the modulation of RRAM conductance by SET/RESET pulses. [ABSTRACT FROM AUTHOR]

To make manipulators fulfill end-effector maintaining tasks, such as writing or drawing tasks in a complex environment, two hybrid multiobjective motion planing schemes, i.e., end-effector posture-maintaining and obstacle avoidance (hybrid PM-OA) schemes are proposed and investigated for wheeled mobile redundant robot manipulators. Specifically, the end-effector posture maintaining, obstacle avoidance, and joint physical limits are considered in a quadratic programming (QP) problem. With these two hybrid PM-OA schemes, the wheeled mobile robot manipulators can maintain its end-effector posture, avoid the obstacle and joint physical limits during executing end-effector tasks. The hybrid PM-OA schemes are finally formulated into a piecewise-linear projection equations (PLPEs) and solved by a recurrent neural network (RNN). Computer simulations are given to substantiate the effectiveness, accuracy, safety, and practicability of the proposed hybrid PM-OA schemes. Comparisons with other schemes and simulations further show that the proposed hybrid PM-OA schemes are more suitable for applications. [ABSTRACT FROM AUTHOR]

Epoxy resin is widely deployed as a high voltage electrical insulation material when compounded with inorganic fillers. However, in the laboratory, the filler prevents visual observation of the long-term degradation known as electrical treeing. To date therefore, much laboratory testing has been conducted on unrepresentative unfilled materials. Here, the impact of micro-sized fillers on the treeing phenomenon in an epoxy system has been explored. Sub-micrometre resolution 3D reconstructions of electrical trees are reported from X-ray computed tomography (XCT) using an advanced ‘pink beam’ synchrotron light source imaging system. The role of filler particles between 1 and 10 μm in size on tree channel propagation is reported. In highly filled materials (30% by weight) a radical change in tree growth behavior is seen, leading to bush tree rather than branch tree growth. The dielectric breakdown time at constant stress was also found to increase as the square root of the filler level. The change in geometry of tree growth may explain the extended life of filled materials in high voltage applications. [ABSTRACT FROM AUTHOR]

Deep reinforcement learning (DRL) has been successfully applied to end-to-end autonomous driving, especially in simulation environments. However, common DRL approaches used in complex autonomous driving scenarios sometimes are unstable or difficult to converge. This paper proposes two approaches to improve the stability of the policy model training with as few manual data as possible. For the first approach, reinforcement learning is combined with imitation learning to train a feature network with a small amount of manual data for parameters initialization. For the second approach, an auxiliary network is added to the reinforcement learning framework, which can leverage the real-time measurement information to deepen the understanding of environment, without any guide of demonstrators. To verify the effectiveness of these two approaches, simulations in image information-based and lidar information-based end-to-end autonomous driving systems are conducted, respectively. These approaches are not only tested in the virtual game world, but also applied in Gazebo, in which we build a 3D world based on the real vehicle model of Ranger XP900 platform, the real 3D obstacle model, and the real motion constraints with inertial characteristics, so as to ensure that the trained end-to-end autonomous driving model is more suitable for the real world. Experimental results show that the performance is increased by over 45% in the virtual game world, and can converge quickly and stably in Gazebo in which previous methods can hardly converge. [ABSTRACT FROM AUTHOR]

This paper reports observations of electrical tree growth by X-ray computed tomography (XCT) in multiple steps. Growth between times of observation has been obtained by subtraction of an image from the subsequent one. Partial discharge (PD) signals were also recorded during tree growth. It is determined that not all PDs support tree extension. By comparing the energy released by PD and the vaporization energy of newly grown tree channels, it is found that only the PDs that reach the tree tips can contribute to the growth of tree length. In such cases, the efficiency of energy conversion was very high, whereas PD resulting in increases in tree channel diameter had a much lower conversion efficiency. Nano XCT images reveal many finer branches along the trunk of the main tree and in the front of the main channel tree tips. It seems that those finer branches led to tree growth, but were not directly associated with local PD. The new imaging techniques have shown great value in resolving the relationship between partial discharge activity and electrical tree growth, providing a platform for deeper understanding of the tree growth process, and interpretation of partial discharge measurements. [ABSTRACT FROM AUTHOR]

Electric tree growth is a key ageing mechanism leading to breakdown of high voltage electrical insulation. Partial discharges (PDs) are invariably associated with electrical tree inception and propagation. In turn, the physical structure of an electrical tree influences the characteristics of partial discharge activity. Interpretation of PD patterns is therefore central to developing an understanding of the tree propagation process, and also to the use of PD patterns as an asset management tool. Our previous research indicates that the phase resolved PD (PRPD) patterns and pulse sequence analysis (PSA) patterns evolve with tree propagation. A method was proposed to estimate the point-on-wave inception and extinction voltages of PDs in tree channels within each power cycle. It was shown that the evolution of PD patterns is a consequence of changes to PD inception and extinction voltages as a tree develops. This paper provides a deterministic model of partial discharge in tree channels. Simulations of PDs in a straight non-conductive tree channel are based on experimental PD inception, extinction and residual voltages. The quantitative simulations reproduce almost all the characteristics of observed PRPD and PSA patterns. It is concluded that PD events are determined by five key parameters: tree structure, applied voltage, PD inception voltage, PD extinction voltage and PD residual voltage. Key parameters estimated by the method, and the models proposed explain PD activity in non-conductive trees. It is suggested that the PSA and PRPD patterns should be discussed together to fully understand the PD events. This model forms a platform for generating robust information for asset managers using PD measurements from high voltage equipment in service. [ABSTRACT FROM AUTHOR]

Partial discharge (PD) characteristics are inherently linked to associated electrical tree growth characteristics. This paper employs both phase-resolved patterns and pulse-sequence analysis (PSA) to study PDs in a short branch-tree grown from a metallic needle tip. It is found that the shape of PSA dV-dV patterns vary with applied voltage; in particular the values of characteristic voltage difference increase with applied voltage. At 10 kV the value of characteristic voltage difference decreases with tree growth, the average number of PDs per cycle increases, the PD magnitudes decrease, and the characteristic wing-like phase-resolved pattern evolves into a turtle-like pattern. This paper proposes a physical model relating PD activity to tree development, considering space charge accumulation inside the tree channel. All key features of experimental results are explained by the model. It is shown that the characteristic voltage differences are determined by the local PD inception and extinction electric fields. These key parameters determining the PD events can be calculated giving a clear physical interpretation of PD measurements, which is invaluable to those using PD as a condition monitoring or asset management tool. [ABSTRACT FROM AUTHOR]

Redundancy resolution is of great importance in the control of manipulators. Among the existing results for handling this issue, the quadratic program approaches, which are capable of optimizing performance indices subject to physical constraints, are widely used. However, the existing quadratic program approaches require exactly knowing all the physical parameters of manipulators, the condition of which may not hold in some practical applications. This fact motivates us to consider the application of adaptive control techniques for simultaneous parameter identification and neural control. However, the inherent nonlinearity and nonsmoothness of the neural model prohibits direct applications of adaptive control to this model and there has been no existing result on adaptive control of robotic arms using projection neural network (PNN) approaches with parameter convergence. Different from conventional treatments in joint angle space, we investigate the problem from the joint speed space and decouple the nonlinear part of the Jacobian matrix from the structural parameters that need to be learnt. Based on the new representation, we establish the first adaptive PNN with online learning for the redundancy resolution of manipulators with unknown physical parameters, which tackles the dilemmas in existing methods. The proposed method is capable of simultaneously optimizing performance indices subject to physical constraints and handling parameter uncertainty. Theoretical results are presented to guarantee the performance of the proposed neural network. Besides, simulations based on a PUMA 560 manipulator with unknown physical parameters together with the comparison with an existing PNN substantiate the efficacy and superiority of the proposed neural network, and verify the theoretical results. [ABSTRACT FROM PUBLISHER]

Electrical tree growth is a precursor to dielectric breakdown in high voltage polymeric insulation. Partial discharge (PD) has a close relationship with electrical tree propagation and can be both used to understand the aging process, and as a diagnostic tool for asset management. In this paper it is shown that PD patterns change through the early stages of tree growth, and consideration of these changes gives insight into the processes of tree growth. Here, trees have been grown in epoxy resin in needle-plane geometries with 2 mm gaps, at 15 kV peak AC. The PD phase-resolved pattern can be regarded as a combination of the well-known turtle-like and wing-like PD patterns. As a tree extends its length, a wing-like pattern is seen and the maximum discharge magnitude has an almost linear relationship with the maximum length of the growing branch. Comparison of the energy released by discharges and the vaporization energy needed for tree growth supports the proposal that the wing-like pattern corresponds to PDs responsible for growth in length of the trees. Implications for mechanisms for tree growth are considered. Results suggest that asset managers may be able to use partial discharge analysis to distinguish different stages of tree growth, providing a valuable prognostic tool for optimising high voltage plant management and replacement. [ABSTRACT FROM PUBLISHER]

A triple-coil pulsed magnet aiming at 100 T (Tesla) was designed and tested at the Wuhan National High Magnetic Field Center. The magnet was powered by a 100 MJ pulse generator, a battery bank, and 25 kV capacitor banks in combination. The magnet failed at 83 T, while the whole structure remained intact. The failure process was analyzed by using the recorded waveforms of the voltage and the current, as well as disassembling and dismantling the failed magnet. It shows that the inner coil broke down first, and then the electrical breakdown occurred between the middle and outer coils. The possible reasons were also discussed. The buckling analysis results show that the inner and middle coils would buckle at 71 T and 83 T, respectively, which were consistent with the experimental results. The cross-ply reinforcement method is proposed to enhance the structural stability of high field magnets. [ABSTRACT FROM AUTHOR]

The synergetic effects of total ionization dose and high-temperature stress of silicon on insulator (SOI) FinFETs (FF) are investigated under ON-state bias condition. The experiments and TCAD simulations are employed to analyze the influence of different trapped-charges in the gate and buried oxide on the threshold voltage and mobility variations on the ON-state current of the devices. Results show that the threshold voltage degradation worsens for n-SOI FinFETs but weakens for the p-SOI FinFETs. Also, the ON-state current of n-SOI FinFETs is significantly enhanced, while the change of p-SOI FinFETs is weak under the synergetic effects. The threshold voltage degradation induced by synergetic effects for both n- and p-type SOI FinFETs can exceed the linear superposition of the two types of stresses, which may be related to the thermal release of trapped charges in the gate oxide. The enhanced degradation in the n-SOI FinFETs compared with the p-SOI FinFETs may result from the shallow energy level electron traps in HfO2. Besides, for both n- and p-type SOI FinFETs, the ON-state current variations under the synergetic effects depend on the competitive process of threshold voltage variation and the mobility fluctuation. [ABSTRACT FROM AUTHOR]

Hyperspectral images (HSIs) contain many levels of spatial structures, and thus, feature extraction techniques have been broadly studied in hyperspectral data processing to mine its structural information while preserving strong edge details. However, spatial characteristics in different scales unavoidably mix and overlap, which makes separation among pixels very difficult and yields misclassification. This article proposes a feature extraction method by using multitask superpixel-based auxiliary learning (MSAL) that consists of two main stages. We first design a novel multitask learning (MTL) model from the perspective of task input by constructing a set of auxiliary feature cubes as inputs. Under the MTL framework, the spatial information and edge information are refined in the superpixel-based auxiliary learning feature extraction stage. Specifically, the auxiliary rolling guidance features are extracted from auxiliary cubes via rolling guidance filter, which can remove the unwanted structure. Then, the spatial feature within superpixels is refined by applying structure density-based weighted operation, and the edge feature is stacked with the superpixel auxiliary features to supply the edge information. Finally, the obtained feature cubes are fed into a pixelwise classifier, and a weighted decision rule is introduced to optimize the initial results by utilizing the inherent relation between auxiliary cubes. Experimental results demonstrate that the MSAL can significantly improve the quality of the extraction feature and outperform other state-of-the-art approaches in terms of classification accuracies. [ABSTRACT FROM AUTHOR]

Electrical trees are widely grown in laboratories from metallic needle electrodes to enable the degradation process to proceed rapidly at voltages of ∼10 kV AC. Here it is shown that cone-shaped air gaps may form at a needle tip, even when they are not visible optically. Using X-ray imaging techniques, samples with and without air gaps up to 55 μm long have been characterized. The presence of air gaps led to partial discharge (PD) patterns resembling void-discharges. Prior to tree initiation, discharge characteristics evolved with increased voltage increments and increased phase lag behind the applied voltage. These changes to PD characteristics reflect a change in the physical state of the air gap. Although samples with air gaps generally initiated trees quicker than those with no gap, PD activities did not appear to play a decisive role in tree initiation times. However, initial voids were found to have major impact on the subsequent tree growth. In samples with persistent pre-tree air-gap PDs, trees grew larger branchy structures with higher PD magnitudes. In addition, the PDs became more asymmetric between positive and negative half-cycles than occurred in samples with no air gaps. It is shown that understanding the interface between the needle and polymer at a micron scale is critical to interpretation of laboratory tree growth experiments. [ABSTRACT FROM AUTHOR]