Abstract: Current terminal devices, such as Personal computers (PCs), diskless workstation, mobile terminals and embedded Internet of things (IoT) devices, are facing several challenges, including limited computing capabilities and storage resources, complex maintenance and management of the software and data, as well as secure service provisioning. Facing these challenges, Cloud computing (CC) and Edge computing (EC) encourage the terminal devices to shift compute-intensive works to the cloud servers or edge nodes. Different from CC and EC, Transparent computing (TC) is a novel user-centric computing paradigm in which the hardware and software are separated in different places. Specificaly, all the software, including Operating systems (OSes), applications programs and management tools, are deposited on the servers and transmitted on demand to terminal devices in a transparent way, while the computation is performed on terminals. In this article, we give a broad survey on transparent computing about the development and current status and compare the differences with CC and EC. To realize the provisioning of the cross-platform service, TC shields hardware differences of heterogeneous devices and builds standardized hardware and software interfaces. From the perspective of security, TC entails high-level security mechanisms and the Meta OS can detect malware and OS-level attacks. Specifically, we first review the development courses of network computing paradigms and the motivation of TC. Then, we present a comparison of several emerging computing paradigms and TC from three perspectives, i.e., virtualization, location for computing, and location for storage. We also show the basic architecture of TC, development process, current status, and key enabling technologies of TC for PCs and lightweight terminals. Lastly, several new challenges and future research directions are indicated to inspire continuous investigations. We believe that this survey can help readers to obtain complete information about TC.
Abstract: This paper proposes a novel image-based method to generate real-time reflections. Our algorithm aims to solve the problem of missing reflections in previous screen space based methods efficiently. Instead of computing reflections in the image space decided by the current eye position, our method conducts most of the reflection calculations in a new image space, which is determined by a carefully chosen camera position at runtime. The major criteria used to choose this optimized camera position is to make its corresponding image space contain as much as possible scene information to generate reflections. Our experimental results indicate that most of the missing reflections problems can be fixed by moving the reflection computations to this new image space. We also prove that the performance gains achieved by executing computations in the new image space can cancel out most of the overhead to generate it.
Abstract: The patients' medical image data are one of the most important data in e-health. Medical image data usually play a crucial role in disease diagnosis and implicate unpredictable potential values for improving diagnostic methods and adjusting diagnostic results. To exploit their incredible potential values, medical images need to be shared among different hospitals, medical institutions and insurance companies and others. But how to securely and effectively share these medical image data becomes a challenging problem. In this paper, we proposed to combine encryption and digital watermark technology to achieve a secure and privacy-preserving medical image sharing method. The QR code image of the concatenation of authoritative diagnosis results and the hash of the original medical image is generated as the watermark image. The Discrete cosine transform (DCT) and Inverse DCT (IDCT) algorithms are utilized to embed the watermark image. As the watermarked medical images are desensitized, they are stored to a smart contract based blockchain, such as Ethereum, to achieve secure and fair sharing between data owners and users. The experimental results show that the proposed method can resist several attacks meanwhile it is efficient in medical image sharing.
Abstract: The key of swaying tree simulation is to capture the essence of the interaction between wind and tree, which has not received much attention from scholars. In this paper, we present a method for simulating swaying tree based on slicing partition. The tree is divided into multiple "slices", and then jointly moved to simulate the tree swaying in wind. By slicing partition, the motion of any slice is independent of the others concurrently, which provides a parallel spatial topology for real-time calculation. Moreover, based on the fact that the wind will blow to a tree partly rather than the whole, partial swayings of the tree are also simulated. The experimental results show the correctness and effectiveness of our method.
Abstract: Aiming at the multi-objective polarity design of Mixed-polarity Reed-Muller (MPRM) circuit, such as small area and low power consumption, an integrated polarity optimization scheme based on improved Multi-objective particle swarm optimization (MOPSO) is proposed. In the Improved MOPSO (IMOPSO) algorithm, particles in the external archive can be actively evolved through self-learning operations to find better circuit polarity. The particles in the population achieve selflearning fractals by comparing the differences between their own states and individuals in external archive to enhance the evolutionary level of the population. A multiobjective decision model of area and power consumption is established according to the characteristics of MPRM circuit. The tabular technique and the IMPOPO algorithm are combined to obtain the Pareto optimal polarity set of the MPRM circuit for area and power consumption. The MCNC Benchmark circuit is used to test the performance of the algorithm. The results verify the effectiveness of the proposed algorithm.
Abstract: The Formation reconfiguration problem plays a crucial role in the implementation of complex tasks for multiple unmanned aerial vehicles, which attracted increasing attention in the past decade. Taking into consideration the control parameters and time discretization of the multi-UAVs in the 3 Dimensional (3-D) space, the formation reconfiguration problem can be formulated as a large-scale combinatorial optimization problem with complex constraints and tight couplings between variables. The problem results in the reduction in efficiency and effectiveness using classic bio-inspired algorithms. In this paper, a formation reconfiguration method based on cooperative coevolutionary algorithm is proposed along with a new decomposition strategy to improve the optimization capability and prevent premature convergence. In the proposed approach, variables of multi-UAV are divided into several sub-groups based on an adaptive grouping strategy. The proposed strategy groups the variables in order to better deal with the tight coupling among them, taking into account the variables' variance and multi-UAVs characteristics of the formation reconfiguration problem. Therefore, each subgroup can adopt the Self-adaptive differential evolution strategy with neighborhood search (SaNSDE) with the aim to optimize the UAV's control inputs using multithreaded programming. SaNSDE contributes to calculating the results in a fully distributed and paralleled manner. Optimal solution is then obtained through cooperation and coordination with all subcomponents. Simulation results based on extreme scenarios adopted by previous researches demonstrate that the proposed algorithm outperformed the existing approaches including Particle swarm optimization (PSO), Differential evolution (DE), and the cooperative coevolution algorithms with different well-known grouping strategies.
Abstract: Genetic algorithms (GAs) serve as a class of powerful tools to search for an effective multicast routing scheme among multiple cluster header nodes, which strongly affects the lifetime of two-tiered Wireless sensor networks (WSNs). This paper proposes a novel Genetic algorithm (GA) with a new crossover mechanism called Leaf crossover for the multicast routing among upper tier nodes in two-tiered WSNs, which outperforms the existing popular tree-based GAs by not requiring the global network link information, encoding/decoding or repair operations. Our simulation study indicates that the proposed algorithm could prolong the lifetime of multicast service, increase the packet delivery ratio as well as converge fast by comparison with existing GAs.
Abstract: ZpZps-additive cyclic codes have been proved to be asymptotically good by Yao and Zhu. For binary Hamming scheme, we introduce a type of Z2(Z2 + uZ2)-additive cyclic codes generated by pairs of polynomials. Let R be the chain ring Z2 + uZ2, where u2=0. The asymptotic rates and relative distances of this class of codes are presented by establishing the relationship between the random Z2R-additive cyclic code and random binary quasi-cyclic code of index 2. We show that Z2R-additive cyclic codes are asymptotically good.
Abstract: By permutation behavior of certain linearized polynomials, the bentness of quadratic vectorial bent functions of the form F (x)=+Trmkt(ckx1+2kt) is investigated, where n=2kt and m|kt with k, t being positive integers. The numerical results show that there exist new quadratic vectorial bent functions obtained up to extended affine equivalence
Abstract: A class of quadratic vectorial bent functions having the form F (x)=Trmn (ax2s1 +1) + Tr1n (bx2s2 +1) is investigated, where n, m, s1, s2 are positive integers and the coefficients a, b belong to the finite field F2n. Through some discussions on the permutation property of certain linearized polynomials over F2n, several classes of quadratic vectorial bent functions are presented for special cases of n, and it is also verified by computer that some vectorial bent functions proposed are extended affine inequivalent to all known quadratic vectorial bent functions.
Abstract: We propose a multi-channel sliced deep Recurrent convolutional neural network (RCNN) with a residual network. We expand the RCNN into a deep neural network. Our proposed model can directly learn to extract bigram features and other features from sentences where other machine learning methods cannot. The experimental results indicate that our model outperforms the traditional methods.
Abstract: Proof of stake (PoS), aiming at replacing Proof of work (PoW) in blockchain consensus, has drawn great attention from academia and industry. We present "Baguena", a novel PoS protocol for public blockchain with high practicality and security. It uses a special designed algorithm with properties of uniqueness and anonymity for leader selection, and uses the longest chain rule for chain selection. Besides, entropy is introduced to prevent manipulation of leader selection process by simulating a random beacon based on Publicly verifiable secret sharing (PVSS) and threshold signature with only a linear number of exponentiations. We analyze the protocol's security by a threat model and design a robust delegation mechanism based on triple Elliptic curve digital signature algorithm (ECDSA) proxy signature. We implement Baguena and evaluate its performance on 100 Amazon EC2 virtual machines simulating 50k users, which shows that Baguena confirms transactions in 2 minutes, achieves 2.16×of Algorand's throughput and 6.95×of Ouroboros' throughput.
Abstract: This paper proposes an improved Empirical mode decomposition (EMD) method piecewise cubic hermite interpolation and mirror extension. Firstly, mirror expansion method is used to expand preliminary signal endpoint; then the piecewise cubic Hermite interpolation algorithm is applied to obtain the envelopes. Finally Intrinsic mode functions (IMFs) are obtained through improved EMD (MPC-HEMD). Through the analysis on simulation signals and Electroencephalogram (EEG) signals, we found that the new method was better than the traditional method.
Abstract: Image segmentation and image decomposition are fundamental problems in image processing. Image decomposition methods for separating images into cartoon and texture components can effectively serve different image processing tasks because different components can be respectively treated in more effective way. However, image decomposition methods are currently simply taken as an independent preprocessing step, and particularly in image segmentation different effects of cartoon and texture components have not been considered. This paper presents a novel simultaneous cartoon-texture image segmentation and image decomposition method to boost the performance of both segmentation and decomposition. We design a fast alternating optimization algorithm to solve the proposed model. Experimental results demonstrate the outstanding performance of the proposed method on both image segmentation and image decomposition.
Abstract: This paper proposes a robust sparse descriptor based on tensor theory by using the spatial and spectral information synthetically, namely the Tensor gradient SIFT (TGSIFT), for Hyperspectral image (HSI). TGSIFT integrates both spatial and spectral information and considers the natural vector feature of HSIs. Based on the HSI Gaussian scale space, a new tensor model for HSI is proposed which takes the vectorial nature of HSI into consideration and preserves all the necessary structural information distributed over all the bands. The TGSIFT descriptor is formed based on the model proposed. Experimental results of HSI matching show that the TGSIFT descriptor achieves better matching performance than other SIFT descriptors under different transformations, including illumination change, sensor noise, image rotation, viewpoint change, and scale change.
Abstract: The paper investigates the hidden relationships among speech samples by applying graph tools. Specifically, we first estimate an applicable graph topology for unstructured speech signals, which can map speech signals into the vertex domain successfully and construct as Speech graph signals (SGSs). On the basis, we define a new graph Fourier transform for SGSs, which can investigate its related graph Fourier analysis. Moreover, we propose a new Graph structure spectral subtraction (GSSS) method for speech enhancement under different noisy environments. Simulation results show that the performance of the GSSS method can be significantly improved than the classical Basic spectral subtraction (BSS) method in terms of the average Segmental signal-tonoise ratio (SSNR), Perceptual evaluation of speech quality (PESQ) and the computational complexity.
Abstract: Though various theoretical results and algorithms have been proposed in one-bit Compressed sensing (1-bit CS), there are few studies on more structured signals, such as block sparse signals. We address the problem of recovering block sparse signals from one-bit measurements. We first propose two recovery schemes, one based on second-order cone programming and the other based on hard thresholding, for common non-adaptively thresholded one-bit measurements. Note that the worst-case error in recovering sparse signals from non-adaptively thresholded one-bit measurements is bounded below by a polynomial of oversampling factor. To break the limit, we introduce a recursive strategy that allows the thresholds in quantization to be adaptive to previous measurements at each iteration. Using the scheme, we propose two iterative algorithms and show that corresponding recovery errors are both exponential functions of the oversampling factor. Several simulations are conducted to reveal the superiority of our methods to existing approaches.
Abstract: Greedy algorithms are widely used for sparse recovery in compressive sensing. Conventional greedy algorithms employ the inner product vector of signal residual and sensing matrix to determine the support, which is based on the assumption that the indexes of the larger-magnitude entries of the inner product vector are more likely to be contained in the correct supports. However, this assumption may be not valid when the number of measurements is not sufficient, leading to the selection of an incorrect support. To improve the accuracy of greedy recovery, we propose a novel greedy algorithm to recover sparse signals from incomplete measurements. The entries of a sparse signal are modelled by the type-II Laplacian prior, such that the k indexes of the correct support are indicated by the largest k variance hyperparameters of the entries. Based on the proposed model, the supports can be recovered by approximately estimating the hyperparameters via the maximum a posteriori process. Simulation results demonstrate that the proposed algorithm outperforms the conventional greedy algorithms in terms of recovery accuracy, and it exhibits satisfactory recovery speed.
Abstract: Accurate harvested energy prediction of the energy harvesting Internet-of-things (IoT) nodes is the basis of the proper power management and should be lowoverhead. A new multi-algorithm fusion framework, which merges results of multiple prediction algorithms to achieve a higher accuracy, has been proposed. A three-algorithm fusion solar radiation predictor was implemented. The experiments using the real solar radiation data show that it improves the percentage prediction error by 10%-26% for different prediction intervals. Its complexity is low enough to run on the embedded systems in real-time.
Abstract: This manuscript presents a new fourquadrant analog multiplier using a recently reported current mode active building block, namely the Fully differential second generation current conveyor (FDCCII). The proposed circuit employs single FDCCII and two NMOSFETs only, thus has simple architecture. It is fully-integrable as no other external passive component has been used. Non-ideal behaviour of the reported configuration has been analysed considering current and voltage tracking errors of the FDCCII. Workability of the derived multiplier is verified with PSPICE (Cadence 16.6) simulations using model parameter of TSMC 0.35μm CMOS process and found to be in close agreement with theoretical anticipations. The static power consumption of the circuit is 0.107mW. The circuit works well with good linearity (nonlinearity error ≤ 0.96%) for the input voltage range of ±0.5V for a supply voltage of ±1V and the output is insensitive to temperature variation. Simulation results show that the -3dB bandwidth of the proposed multiplier is 20.67MHz and the output referred noise is less than 9nV/√Hz at 1kΩ load condition. MonteCarlo analysis has also been performed for the proposed configuration. The applicability of the reported multiplier as amplitude modulator, squarer, and frequency doubler are also demonstrated.
Abstract: Pantograph arcing caused by Off-line of pantograph-catenary (OLPC) will generate Electromagnetic disturbance (EMD), which will affect train control system and communication system. This paper proposes a simplified model of the train and pantograph to investigate propagation characteristics of EMD from OLPC in the viaduct scenario, containing attenuation law in transverse and longitudinal directions. As a result, the electric field strength of EMD from OLPC increases with distance in transverse direction from 10m to 30m in the viaduct scenario, which is different from the ground scenario. The fitting formulas are found to research the attenuation law at different frequency points. We get the conclusion that the limit and measurement distance is not applicable for the viaduct scenario in current standard. To prove this, measurement in the viaduct scenario is completed, and the plausible evidence to explain the result is discussed. These results are useful in the measurement of radiated emissions, on-board equipment layout and the research of standard in high-speed railway.
Abstract: The three Dimensional (3D) visualization of time-varying Electromagnetic (EM) data in antenna devices can be used to evaluate various parameters, such as the quality of beams and the power performance of antennas, for optimizing the antenna design. Compared with the single volume data, the time-varying data has characteristics of strong continuity, large scale and multi-variable. It is challenging. We proposed a real-time dynamic rendering method using Compute unified device architecture (CUDA)-based volume rendering algorithm. We designed the opacity transfer function as multiple trapezoid, which represents each energy band, and provided a fusion strategy for a horn-like surface and a EM volume model that considers the contribution of both intersection points of surface with viewing ray and sample points of the volume. We also proposed a framework for time-varying 3D EM datasets. Two assessment tests were conducted by both experts and students. Results showed that our 3D visualization method were effective.
Abstract: In a three-phase power convertor system, phase current reconstruction technique using a DC-link resistance can reduce the cost and weight of the diving system. The three-phase current without DC offset in normal condition is introduced in this paper firstly. Based on this premise, the DC-link current offset is considered and the logical relationship between threephase voltage, three-phase current, and DC offset are analyzed and depicted. The DC offset introduced during the sampling processes a distortion into phase current, which is composed of fundamental and harmonic frequency based on FFT analysis. The fundamental component of the distortion shares the same phase angle with counterpart voltage, which distorts power factor and amplitude of three-phase currents. The harmonic components of the distortion can be transformed into 6th harmonic components in synchronous dq axes and undermine the performance of the controlling system. After that, a practical method is proposed to solve this problem, namely sampling the DC offset signal during the period of the zero voltage vectors. An analogy sample circuit is designed and tested. This method can avoid the DC offset from the source. Finally, its effectiveness is verified via simulation and experiment.