Since the landmark work of R. E. Kalman in the 1960s, considerable efforts have been devoted to time series state space models for a large variety of dynamic estimation problems. In particular, parametric filters that seek analytical estimates based on a closed-form Markov–Bayes recursion, e.g., recursion from a Gaussian or Gaussian mixture (GM) prior to a Gaussian/GM posterior (termed ‘Gaussian conjugacy’ in this paper), form the backbone for a general time series filter design. Due to challenges arising from nonlinearity, multimodality (including target maneuver), intractable uncertainties (such as unknown inputs and/or non-Gaussian noises) and constraints (including circular quantities), etc., new theories, algorithms, and technologies have been developed continuously to maintain such a conjugacy, or to approximate it as close as possible. They had contributed in large part to the prospective developments of time series parametric filters in the last six decades. In this paper, we review the state of the art in distinctive categories and highlight some insights that may otherwise be easily overlooked. In particular, specific attention is paid to nonlinear systems with an informative observation, multimodal systems including Gaussian mixture posterior and maneuvers, and intractable unknown inputs and constraints, to fill some gaps in existing reviews and surveys. In addition, we provide some new thoughts on alternatives to the first-order Markov transition model and on filter evaluation with regard to computing complexity.
With supercomputers developing towards exascale, the number of compute cores increases dramatically, making more complex and larger-scale applications possible. The input/output (I/O) requirements of large-scale applications, workflow applications, and their checkpointing include substantial bandwidth and an extremely low latency, posing a serious challenge to high performance computing (HPC) storage systems. Current hard disk drive (HDD) based underlying storage systems are becoming more and more incompetent to meet the requirements of next-generation exascale supercomputers. To rise to the challenge, we propose a hierarchical hybrid storage system, on-line and near-line file system (ONFS). It leverages dynamic random access memory (DRAM) and solid state drive (SSD) in compute nodes, and HDD in storage servers to build a three-level storage system in a unified namespace. It supports portable operating system interface (POSIX) semantics, and provides high bandwidth, low latency, and huge storage capacity. In this paper, we present the technical details on distributed metadata management, the strategy of memory borrow and return, data consistency, parallel access control, and mechanisms guiding downward and upward migration in ONFS. We implement an ONFS prototype on the TH-1A supercomputer, and conduct experiments to test its I/O performance and scalability. The results show that the bandwidths of single-thread and multi-thread ‘read’/‘write’ are 6-fold and 5-fold better than HDD-based Lustre, respectively. The I/O bandwidth of data-intensive applications in ONFS can be 6.35 times that in Lustre.
Public verification of data integrity is crucial for promoting the serviceability of cloud storage systems. Recently, Tan and Jia (2014) proposed an identity-based public verification (NaEPASC) protocol for cloud data to simplify key management and alleviate the burden of check tasks. They claimed that NaEPASC enables a thirdparty auditor (TPA) to verify the integrity of outsourced data with high efficiency and security in a cloud computing environment. However, in this paper, we pinpoint that NaEPASC is vulnerable to the signature forgery attack in the setup phase; i.e., a malicious cloud server can forge a valid signature for an arbitrary data block by using two correct signatures. Moreover, we demonstrate that NaEPASC is subject to data privacy threats in the challenge phase; i.e., an external attacker acting as a TPA can reveal the content of outsourced data. The analysis shows that NaEPASC is not secure in the data verification process. Therefore, our work is helpful for cryptographers and engineers to design and implement more secure and efficient identity-based public auditing schemes for cloud storage.
Cascaded regression has been recently applied to reconstruct 3D faces from single 2D images directly in shape space, and has achieved state-of-the-art performance. We investigate thoroughly such cascaded regression based 3D face reconstruction approaches from four perspectives that are not well been studied: (1) the impact of the number of 2D landmarks; (2) the impact of the number of 3D vertices; (3) the way of using standalone automated landmark detection methods; (4) the convergence property. To answer these questions, a simplified cascaded regression based 3D face reconstruction method is devised. This can be integrated with standalone automated landmark detection methods and reconstruct 3D face shapes that have the same pose and expression as the input face images, rather than normalized pose and expression. An effective training method is also proposed by disturbing the automatically detected landmarks. Comprehensive evaluation experiments have been carried out to compare to other 3D face reconstruction methods. The results not only deepen the understanding of cascaded regression based 3D face reconstruction approaches, but also prove the effectiveness of the proposed method.
The nodes number of the hidden layer in a deep learning network is quite difficult to determine with traditional methods. To solve this problem, an improved Kullback-Leibler divergence sparse autoencoder (KL-SAE) is proposed in this paper, which can be applied to battle damage assessment (BDA). This method can select automatically the hidden layer feature which contributes most to data reconstruction, and abandon the hidden layer feature which contributes least. Therefore, the structure of the network can be modified. In addition, the method can select automatically hidden layer feature without loss of the network prediction accuracy and increase the computation speed. Experiments on University of California-Irvine (UCI) data sets and BDA for battle damage data demonstrate that the method outperforms other reference data-driven methods. The following results can be found from this paper. First, the improved KL-SAE regression network can guarantee the prediction accuracy and increase the speed of training networks and prediction. Second, the proposed network can select automatically hidden layer effective feature and modify the structure of the network by optimizing the nodes number of the hidden layer.
Recently, low-dose computed tomography (CT) has become highly desirable because of the growing concern for the potential risks of excessive radiation. For low-dose CT imaging, it is a significant challenge to guarantee image quality while reducing radiation dosage. Compared with classical filtered backprojection algorithms, compressed sensing-based iterative reconstruction has achieved excellent imaging performance, but its clinical application is hindered due to its computational inefficiency. To promote low-dose CT imaging, we propose a promising reconstruction scheme which combines total-variation minimization and sparse dictionary learning to enhance the reconstruction performance, and properly schedule them with an adaptive iteration stopping strategy to boost the reconstruction speed. Experiments conducted on a digital phantom and a physical phantom demonstrate a superior performance of our method over other methods in terms of image quality and computational efficiency, which validates its potential for low-dose CT imaging.
We propose an optimal approach to solve the problem of multi-degree reduction of C-Bézier surfaces in the norm L2 with prescribed constraints. The control points of the degree-reduced C-Bézier surfaces can be explicitly obtained by using a matrix operation that is based on the transfer matrix of the C-Bézier basis. With prescribed boundary constraints, this method can be applied to piecewise continuous patches or to a single patch with the combination of surface subdivision. The resulting piecewise approximating patches are globally G1 continuous. Finally, numerical examples are presented to show the effectiveness of the method.
Moving target defense (MTD) is a novel way to alter the asymmetric situation of attacks and defenses, and a lot of MTD studies have been carried out recently. However, relevant analysis for the defense mechanism of the MTD technology is still absent. In this paper, we analyze the defense mechanism of MTD technology in two dimensions. First, we present a new defense model named MP2R to describe the proactivity and effect of MTD technology intuitively. Second, we use the incomplete information dynamic game theory to verify the proactivity and effect of MTD technology. Specifically, we model the interaction between a defender who equips a server with different types of MTD techniques and a visitor who can be a user or an attacker, and analyze the equilibria and their conditions for these models. Then, we take an existing incomplete information dynamic game model for traditional defense and its equilibrium result as baseline for comparison, to validate the proactivity and effect of MTD technology. We also identify the factors that will influence the proactivity and effectiveness of the MTD approaches. This work gives theoretical support for understanding the defense process and defense mechanism of MTD technology and provides suggestions to improve the effectiveness of MTD approaches.
Mathematical models are disappointing due to uneven distribution of the air gap magnetic field and significant unmodeled dynamics in magnetic bearing systems. The effectiveness of control deteriorates based on an inaccurate mathematical model, creating slow response speed and high jitter. To solve these problems, a model-free adaptive control (MFAC) scheme is proposed for a three-degree-of-freedom hybrid magnetic bearing (3-DoF HMB) control system. The scheme for 3-DoF HMB depends only on the control current and the objective balanced position, and it does not involve any model information. The design process of a parameter estimation algorithm is model-free, based directly on pseudo-partial-derivative (PPD) derived online from the input and output data information. The rotor start-of-suspension position of the HMB is regulated by auxiliary bearings with different inner diameters, and two kinds of operation situations (linear and nonlinear areas) are present to analyze the validity of MFAC in detail. Both simulations and experiments demonstrate that the proposed MFAC scheme handles the 3-DoF HMB control system with start-of-suspension response speed, smaller steady state error, and higher stability.
The DC microgrid is connected to the AC utility by parallel bidirectional power converters (BPCs) to import/export large power, whose control directly affects the performance of the grid-connected DC microgrid. Much work has focused on the hierarchical control of the DC, AC, and hybrid microgrids, but little has considered the hierarchical control of multiple parallel BPCs that directly connect the DC microgrid to the AC utility. In this paper, we propose a hierarchical control for parallel BPCs of a grid-connected DC microgrid. To suppress the potential zero-sequence circulating current in the AC side among the parallel BPCs and realize feedback linearization of the voltage control, a d-q-0 control scheme instead of a conventional d-q control scheme is proposed in the inner current loop, and the square of the DC voltage is adopted in the inner voltage loop. DC side droop control is applied to realize DC current sharing among multiple BPCs at the primary control level, and this induces DC bus voltage deviation. The quantified relationship between the current sharing error and DC voltage deviation is derived, indicating that there is a trade-off between the DC voltage deviation and current sharing error. To eliminate the current sharing error and DC voltage deviation simultaneously, slope-adjusting and voltage-shifting approaches are adopted at the secondary control level. The proposed tertiary control realizes precise active and reactive power exchange through parallel BPCs for economical operation. The proposed hierarchical control is applied for parallel BPCs of a grid-connected DC microgrid and can operate coordinately with the control for controllable/uncontrollable distributional generation. The effectiveness of the proposed control method is verified by corresponding simulation tests based on Matlab/Simulink, and the performance of the hierarchical control is evaluated for practical applications.
Ultra-wideband frequency modulated continuous wave (FMCW) radar has the ability to achieve high-range resolution. Combined with the inverse synthetic aperture technique, high azimuth resolution can be realized under a large rotation angle. However, the range-azimuth coupling problem seriously restricts the inverse synthetic aperture radar (ISAR) imaging performance. Based on the turntable model, traditional match-filter-based, range Doppler algorithms (RDAs) and the back projection algorithm (BPA) are investigated. To eliminate the sidelobe effects of traditional algorithms, compressed sensing (CS) is preferred. Considering the block structure of a signal at high resolution, a block-sparsity adaptive matching pursuit algorithm (BSAMP) is proposed. By matching pursuit and backtracking, a signal with unknown sparsity can be recovered accurately by updating the support set iteratively. Finally, several experiments are conducted. In comparison with other algorithms, the results from processing the simulation data, some simple targets, and a complex target indicate the effectiveness and superiority of the proposed algorithm.
Considerable progress has been made recently in the development of techniques to determine exactly two-point resistances in networks of various topologies. In particular, a general resistance formula of a non-regular m×n resistor network with an arbitrary boundary is determined by the recursion-transform (RT) method. However, research on the complex impedance network is more difficult than that on the resistor network, and it is a problem worthy of study since the equivalent impedance has many different properties from equivalent resistance. In this study, the equivalent impedance of a non-regular m×n RLC network with an arbitrary boundary is studied based on the resistance formula, and the oscillation characteristics and resonance properties of the equivalent impedance are discovered. In the RLC network, it is found that our formula leads to the occurrence of resonances at the boundary condition holding a series of specific values with an external alternating current source. This curious result suggests the possibility of practical applications of our formula to resonant circuits.
Acquisition of accurate channel state information (CSI) at transmitters results in a huge pilot overhead in massive multiple input multiple output (MIMO) systems due to the large number of antennas in the base station (BS). To reduce the overwhelming pilot overhead in such systems, a structured joint channel estimation scheme employing compressed sensing (CS) theory is proposed. Specifically, the channel sparsity in the angular domain due to the practical scattering environment is analyzed, where common sparsity and individual sparsity structures among geographically neighboring users exist in multi-user massive MIMO systems. Then, by equipping each user with multiple antennas, the pilot overhead can be alleviated in the framework of CS and the channel estimation quality can be improved. Moreover, a structured joint matching pursuit (SJMP) algorithm at the BS is proposed to jointly estimate the channel of users with reduced pilot overhead. Furthermore, the probability upper bound of common support recovery and the upper bound of channel estimation quality using the proposed SJMP algorithm are derived. Simulation results demonstrate that the proposed SJMP algorithm can achieve a higher system performance than those of existing algorithms in terms of pilot overhead and achievable rate.
We propose a novel channel model for massive multiple-input multiple-out (MIMO) communication systems that incorporate the spherical wave-front assumption and non-stationary properties of clusters on both the array and time axes. Because of the large dimension of the antenna array in massive MIMO systems, the spherical wave-front is assumed to characterize near-field effects resulting in angle of arrival (AoA) shifts and Doppler frequency variations on the antenna array. Additionally, a novel visibility region method is proposed to capture the non-stationary properties of clusters at the receiver side. Combined with the birth-death process, a novel cluster evolution algorithm is proposed. The impacts of cluster evolution and the spherical wave-front assumption on the statistical properties of the channel model are investigated. Meanwhile, corresponding to the theoretical model, a simulation model with a finite number of rays that capture channel characteristics as accurately as possible is proposed. Finally, numerical analysis shows that our proposed non-stationary channel model is effective in capturing the characteristics of a massive MIMO channel.