Search

Existing deep reinforcement learning (DRL) based methods for solving the capacitated vehicle routing problem (CVRP) intrinsically cope with homogeneous vehicle fleet, in which the fleet is assumed as repetitions of a single vehicle. Hence, their key to construct a solution solely lies in the selection of the next node (customer) to visit excluding the selection of vehicle. However, vehicles in real-world scenarios are likely to be heterogeneous with different characteristics that affect their capacity (or travel speed), rendering existing DRL methods less effective. In this paper, we tackle heterogeneous CVRP (HCVRP), where vehicles are mainly characterized by different capacities. We consider both min-max and min-sum objectives for HCVRP, which aim to minimize the longest or total travel time of the vehicle(s) in the fleet. To solve those problems, we propose a DRL method based on the attention mechanism with a vehicle selection decoder accounting for the heterogeneous fleet constraint and a node selection decoder accounting for the route construction, which learns to construct a solution by automatically selecting both a vehicle and a node for this vehicle at each step. Experimental results based on randomly generated instances show that, with desirable generalization to various problem sizes, our method outperforms the state-of-the-art DRL method and most of the conventional heuristics, and also delivers competitive performance against the state-of-the-art heuristic method, i.e., SISR. Additionally, the results of extended experiments demonstrate that our method is also able to solve CVRPLib instances with satisfactory performance., Comment: This paper has been accepted at IEEE Transactions on Cybernetics

In recent years, impressive performance of deep learning technology has been recognized in synthetic aperture radar (SAR) automatic target recognition (ATR). Since a large amount of annotated data are required in this technique, it poses a trenchant challenge to the issue of obtaining a high recognition rate through less labeled data. To overcome this problem, inspired by the contrastive learning, we proposed a novel framework named batch instance discrimination and feature clustering (BIDFC). In this framework, different from that of the objective of general contrastive learning methods, embedding distance between samples should be moderate because of the high similarity between samples in the SAR images. Consequently, our flexible framework is equipped with adjustable distance between embedding, which we term as weakly contrastive learning. Technically, instance labels are assigned to the unlabeled data in per batch, and random augmentation and training are performed few times on these augmented data. Meanwhile, a novel dynamic-weighted variance loss (DWV loss) function is also posed to cluster the embedding of enhanced versions for each sample. The experimental results on the moving and stationary target acquisition and recognition (MSTAR) database indicate a 91.25% classification accuracy of our method fine-tuned on only 3.13% training data. Even though a linear evaluation is performed on the same training data, the accuracy can still reach 90.13%. We also verified the effectiveness of BIDFC in OpenSarShip database, indicating that our method can be generalized to other data sets. Our code is available at: https://github.com/Wenlve-Zhou/BIDFC-master.

Video synthetic aperture radar (ViSAR) is gaining increasing attention in the remote sensing area due to its advantages in continuous observation compared with conventional synthetic aperture radar (SAR). Generally, the polar format algorithm (PFA) is utilized to generate the ViSAR frames considering both processing efficiency and imaging quality. However, the changes in squint angles for different ViSAR frames will cause variations in target positions when the classic fixed-parameter PFA is utilized. Hence, an image registration step is typically needed, which greatly increases the computational load. Meanwhile, the azimuth resolution also deteriorates rapidly for the ViSAR frames as the squint angle becomes larger. In order to solve the aforementioned problems, a parameter-adjusting autoregistration PFA (PAAR-PFA) is proposed for the ViSAR. By adjusting system parameters, such as carrier frequency, pulsewidth, and sampling frequency, according to the azimuth sampling positions, PAAR-PFA can achieve autoregistration for ViSAR frames without range interpolation and geometric correction in fixed-parameter PFA, which significantly improves the processing efficiency. At the same time, since the azimuth sampling number of frame data is adjusted with the squint angle, the variation range of the azimuth resolution under different squint angles is considerably reduced. Point target and extended target simulations confirm the feasibility of the proposed algorithm.

Phase noise reduction is one of the key steps for synthetic aperture radar interferometry data processing. In this article, a novel phase filtering method is proposed. The main innovation and contribution of this research is to 1) incorporate local fringe frequency (LFF) compensation technique into the nonlocal phase filtering method to include more independent and identically distributed samples for filtering; 2) modify the nonlocal phase filter from three aspects: 1) executing nonlocal filtering in the complex domain of the residual phase to avoid gray jumps in phase, 2) adaptively calculating the smoothing parameter based on the LFF and the coherence coefficient, and 3) using the integral image in similarity calculation to improve the efficiency; 3) perform Goldstein filter in high coherence areas to reduce the computation expense. Experiments based on both simulated and real data have shown that the proposed method has achieved a better performance in terms of both noise reduction and edge preservation than some existing phase filtering methods.

For a graph, the label numbers of points and edges can be mapped to, and the label value of any edge and its associated vertices is constant, then the graph is said to have edge-magic total labeling. This paper first obtains undirected simple connected graphs within sixteen points, and designs a new algorithm that can be used to determine whether the graph has edge-magic total labeling. Secondly, the labeling results of undirected simple connected graphs within 16 points are obtained, and several theorems are summarized and proved.

Aiming at the problems of low mining efficiency, strong subjectivity and too many association relations generated by classical mining algorithms, a novel algorithm for association rule mining of high-dimensional data is designed in terms of both sample selection and association rule generation. The algorithm reduces the impact of weak samples at the beginning of mining by calculating the distribution coefficients and deletion thresholds of the samples and synthesizing custom support to double screen the samples at the first reading of the dataset. When generating frequent items and association rules, the algorithm mines information in a sample relationship table and sample the full relationship combination mode, which reduces the complexity and resource consumption of the mining process. The experimental results show that the number of frequent items and association rules mined by the Marc algorithm is significantly reduced, and the mining efficiency and memory consumption of the Marc algorithm are better than those of the Apriori, FP growth and Eclat algorithms. The higher the dimension is, the larger the data set is, and the more obvious the advantage is. The accuracy of the Marc algorithm for mining frequent items and association relations is 100%.

Visible light communication based on display and camera can simultaneously provide the functionalities of display and invisible information transmission. There are widely potential applications in information broadcasting, stealth advertising, etc. In this paper, a real-time display camera communication system based on LED displays and smartphones is proposed. To solve the rate mismatch problem and the symbol unsynchronization problem between the LED display panel with a relatively high frame rate and the smartphones’ camera with a relatively low frame rate, we propose an Alternate Bit-flipping Repeat Coding scheme for adjustable frame rate matching and symbol synchronization with the advantage of eliminating flickers perceivable to human eyes. Fast image processing algorithms are also designed to decrease the computational complexity of extracting information from the captured video frames in real-time. An experimental system was implemented on a 16×16 points LED display panel with a refresh rate of 150 fps and a commercial Android smartphone with an 8 megapixels camera working at 30 fps and a self-developed Android APP. The experimental results show that the real-time LED display camera communication system can achieve a data transmission rate of 30 bps for one LED display point and a maximum distance of 80 cm, and the computational time on the Android smartphone for data receiving is only 11.827 ms.

Recently, there is an emerging trend to apply deep reinforcement learning to solve the vehicle routing problem (VRP), where a learnt policy governs the selection of next node for visiting. However, existing methods could not handle well the pairing and precedence relationships in the pickup and delivery problem (PDP), which is a representative variant of VRP. To address this challenging issue, we leverage a novel neural network integrated with a heterogeneous attention mechanism to empower the policy in deep reinforcement learning to automatically select the nodes. In particular, the heterogeneous attention mechanism specifically prescribes attentions for each role of the nodes while taking into account the precedence constraint, i.e., the pickup node must precede the pairing delivery node. Further integrated with a masking scheme, the learnt policy is expected to find higher-quality solutions for solving PDP. Extensive experimental results show that our method outperforms the state-of-the-art heuristic and deep learning model, respectively, and generalizes well to different distributions and problem sizes., This paper has been accepted at IEEE Transactions on Intelligent Transportation Systems

A method is proposed for visualizing damage information of traditional buildings in Chinese village. Firstly, we label the buildings one by one in experimental area. Secondly, the texture and spectral feature parameters of each building are calculated. For each single building, parameters that are positively correlated with the level of damage are normalized to form a parameter vector, and the module was calculated subsequently. Finally, mapping parameter vector and module to RGB color space, and the distribution, damage type and level of each buildings can be visualized. Experimental results show that this method can realize recognizing damage buildings in high-spatial-resolution imagery, and get initial understanding of buildings damage level. In summary, this method can effectively and rapidly realize the extraction of damaged buildings information with high resolution images.

The traditional system of video surveillance interworking improves the efficiency, resources utilization rates and management ability through emerging technology, such as: AI, Big data, etc. However, while video surveillance system interworking, the problem of video surveillance data leakage, data authenticity cannot be verified still exists. Blockchain is a fundamental trust techniques, which can solve the problem of data not be trusted, data has been tampered, etc. This paper research and proposes an application model of video surveillance system interworking based on blockchain (BVSSI) and shows several application scenarios based BVSSI.

Based on the graph generation algorithm, we propose a novel algorithm to solve the problem of D(β)-vertices sum distinguishing edge coloring of some kinds of special graphs. Our main contributions to solving this graph coloring problem are: firstly, we generate the adjacent matrices of the following graphs; secondly, we input these matrices to our designed algorithm; thirdly, the algorithm will output the corresponding D(β)-vertices sum distinguishing edge coloring matrices; fourthly, through analyzing of the results obtained by our algorithm, we conclude that regular chromatic number are found in some of the graphs involved. We think that our proposed algorithm is innovative and can be adopted by other researchers.

Image fusion method is an algorithm that synthesizes images from different sources into a new image. The advantage of image fusion technology is that different image information presented by multi-source sensors can complement and improve each other to extract the effective information in the source image. It can also obtain a more effective and richer image with description information. According to the technical methods of image fusion, an improved image fusion method for fast non-downsampled contourlet transform (NSCT) was selected for handheld. The source infrared image and source low-light image collected by the observer were image fusion. In this method, the scale filter and the directional filter were convolved respectively in the composition of image decomposition and reconstruction to reduce the number of iterations of image decomposition and reconstruction, improve the efficiency of algorithm operation, and ensure that the system worked in real time.

Advertising posters are a commonly used form of information presentation to promote a product. Producing advertising posters often takes much time and effort of designers when confronted with abundant choices of design elements and layouts. This paper presents Vinci, an intelligent system that supports the automatic generation of advertising posters. Given the user-specified product image and taglines, Vinci uses a deep generative model to match the product image with a set of design elements and layouts for generating an aesthetic poster. The system also integrates online editing-feedback that supports users in editing the posters and updating the generated results with their design preference. Through a series of user studies and a Turing test, we found that Vinci can generate posters as good as human designers and that the online editing-feedback improves the efficiency in poster modification.

Data videos are a genre of narrative visualization that communicates stories by combining data visualization and motion graphics. While data videos are increasingly gaining popularity, few systematic reviews or structured analyses exist for their design. In this work, we introduce a design space for animated visual narratives in data videos. The design space combines a dimension for animation techniques that are frequently used to facilitate data communication with one for visual narrative strategies served by such animation techniques to support story presentation. We derived our design space from the analysis of 82 high-quality data videos collected from online sources. We conducted a workshop with 20 participants to evaluate the effectiveness of our design space. Qualitative and quantitative feedback suggested that our design space is inspirational and useful for designing and creating data videos.

In recent years, synthetic aperture radar (SAR) has played an increasingly important role in the military and civil fields. Since the SAR image reflects the scattering characteristics of the target， it is of great significance to achieve multi-angle fusion of the target. However, there is a problem of angular loss in real SAR images. Through the electromagnetic simulation method, SAR images of 0–360° can be obtained, but the similarity to real images is low. Here, the authors combine electromagnetic simulation with conditional generative adversarial networks (cGANs). The image obtained by the electromagnetic simulation is taken as the input of the cGANs, and then the generator generates photorealistic SAR images containing the label information. Thereby, authors’ method complement the missing angles in the real SAR image dataset. Finally, they qualitatively and quantitatively evaluated the synthetic images generated through their model to verify the quality of the dataset.

During the detection and tracking of high-speed moving target by radar, the coherent accumulation and the energy accumulation of target echo signal are poor. In order to solve this problem, this study discusses speed estimation based on the Hough transform, correction across distance of radar echo signal and pre-tracking detection method based on dynamic programming. The authors propose an energy accumulation method of target echo based on speed estimation. Using this method, speed estimation accuracy reaches 98.87%, signal-to-noise ratio increases by 10 dB and the goal amplitude is 4.81 times more than the original amplitude. Simulation results show that this method can improve the energy accumulation effectiveness and improve detection and tracking ability of high-speed moving target by radar.

Wireless communication is vulnerable to malicious jamming due to its inherent broadcast characteristics. With the development of intelligent technology, the jammer can actively adjust the jamming strategy according to the feedback of the jamming effect, so as to achieve intelligent jamming with lower consumption and higher efficiency. The existence of intelligent jamming poses more serious challenges to the reliable transmission capacity of wireless communication system. In this paper, an intelligent anti-jamming communication method based on game learning is proposed. The communication confrontation between jammer and user is modeled as Stackelberg game, and intelligent anti-jamming decision-making in frequency domain is realized with the help of reinforcement learning algorithm, so as to improve the reliability of wireless transmission and realize intelligent anti-jamming communication in the jamming environment. The simulation results show that the proposed scheme can obtain stable policies, and the proposed anti-jamming communication method has higher anti-jamming performance than the conventional method.

The distribution of urban functional space is an important factor to measure the development of a city. Its reasonable layout plays an important role in the development of urban economy and the optimization of urban spatial pattern. This paper is based on the third ring road of Nanning city Based on POI data, the functional land is divided into five categories: life service, business and finance, public service, leisure and entertainment, and residence. By using kernel density analysis, frequency density analysis and standard deviation ellipse analysis, the urban function and distribution characteristics of the Third Ring Road area in Nanning city are identified and analyzed. The results show that: 1. The overall spatial layout is affected by traffic and urban master plan Under the influence of other factors, the spatial pattern of “circle layer” appears. Taking one ring as the core, it has obvious central agglomeration. 2. The trend of urban development shows an obvious pattern of prosperity in the West and decline in the East.

It is a hotspot problem to quickly extract valuable information and knowledge hidden in the complex, different types, fuzzy and huge amount of space-time trajectory data. In the space-time trajectory data clustering method, according to the existing deficiencies of the classical K-means algorithm, the mathematical distance method and effective iteration method are used to select the initial clustering center to optimize the K-means algorithm, which improves the accuracy and efficiency of the algorithm. Based on MATLAB experimental simulation platform, the comparison experiments between the classical algorithm and the optimized algorithm, the applicability test of the performance test, and the comparison test with the classical algorithm were designed. The experimental results show that the optimized K-means randomly selected initial clustering center is more accurate, which can avoid the drawbacks caused by randomly selected initial clustering center to a certain extent and has better clustering effect on sample data, and at the same time avoid the K-means clustering algorithm falling into the dilemma of local optimal solution in the clustering process.

Three-dimensional light field display, which can provide human with natural 3D images in the way of humans observing the real world without any glasses, has attracted more and more attentions in recent years. However, most of studies focus on the performance improvement of 3D light field display, and there is rarely work on the performance evaluation of 3D light field display. In this paper, a quantitative performance evaluation method of the 3D light field display based on cognitive measurement of the observer with a specifically designed mental rotation task is proposed. The procedure of the performance evaluation experiment design is demonstrated, some pairs of complex images with either same structure or mirror structure are displayed, and the participants are required to distinguish whether the displayed images are the same. Three different control experiments including 2D display with fixed viewpoint, 3D light field display with fixed viewpoint and 3D light field display with a free view are designed. 45 participants which are divided into three groups are assigned to take part in three difference control experiments, and the rate of mission success and the completion time are evaluated. By comparing the results on different experimental conditions and using a statistic analysis method called T-test, the superiority of 3D light field display is proved quantitatively.

In this paper, an adjacent vertex sum reducible edge coloring algorithm (AVSREC) is designed, which can be applied to graphic cryptography. Through the user testing experiment of the identity authentication prototype system, it is verified that the graph password based on the adjacent vertex sum reducible edge coloring algorithm can better prevent shoulder peeping attack and improve the security effectively, and its usability is better than text password and general graphic password.

The distance label is a label model derived from the frequency allocation problem. This article mainly studies the L(2,1)-label in the label. L(2,1)-label refers to the use of non-negative integers to label the vertices of the graph G. The goal is to make the label values of adjacent vertices differ by at least 2 and the label values of vertices with a distance of 2 are different. This paper proposes a recursive backtracking greedy algorithm based on Griggs’ conjecture, and also gives the algorithm’s execution steps, analysis and test results. The experimental results show that for a unicyclic graph containing only one maximum degree point, its L(2,1)-label number is $\Delta$+1. In addition, the L(2,1)-label number characteristics of the unicyclic graph are given and expressed in the form of forbidden subgraphs.

Let G be a simple connected graph with p vertices and q edges, there exists a bijective function f from $V(G)UE(G)$ to the integers 1,2,…, $p+q$, such that the sum of any edge label and its incident vertex labels is equal to a constant k, f is called an edge-magic total labeling(EMTL for short) while k is magic constant. In this paper, an edge-magic total labeling algorithm (EMTL algorithm for short) is designed, which has a lot of graphics processing capabilities. Several complex research problems in some open literatures can be solved easily and effectively.

Aiming at the problem that the training difficulty of neural network model increases with the increase of the number of network layers due to the disappearance of gradient, a method by combining residual U-Net and conditional random field for water body extraction from hyperspectral remote sensing image is proposed. A deeper model based on the U-Net semantic model by adding residual units is constructed, and then a conditional random field is connected after the output of the model to further optimize the output category and refine the image segmentation results. As experimental data, the Zhuhai-1 hyperspectral satellite image is selected to training and validation for the proposed method. The experimental results show that, compared with the water index method, support vector machine, U-Net semantic model and residual U-Net model, the proposed method has the best effect in water body extraction from the hyperspectral remote sensing image.

The Electronic Medical Record (EMR) is a kind of data, which is highly sensitive and worthy of biomedical research. The storage security, integrity, and data sharing of EMRs need further research and discussion. Blockchain is a block-chain storage structure, which has the characteristics of tamper-proof, traceability, and high sharing. Although blockchain technology provides a new solution for the storage and data sharing of EMRs, the research on user-oriented data query is not comprehensive. Traditional SQL database opens its data access, so the data stored in it has easily been tampered with, which is not suitable for storing sensitive medical data. Based on the characteristics of blockchain and traditional databases, we design a model with high security and easy retrieval for the storing and querying the EMRs. In the case of ensuring the safety of medical data storage, it is convenient for users to query and retrieve.

The paper proposes a digital cash (Dcash) online payment system based on exchange center that represents central bank to manage and monitor issuance, payment and refund of Dcash. Difference Dcash could be exchanged when its issuer registered at the exchange center. Then it introduces architecture of the system and a series of transaction procedures. This paper models the payment activity using Color Petri Nets. At last, by analyzing the state space reports of the model, we can verify and check this system and get the results that the flow charts of these entities are controllable, realizable and feasible.

This paper proposes a novel track initiation algorithm based on agglomerated hierarchical clustering and correlation coefficients. The algorithm firstly using the agglomerative hierarchical clustering algorithm to classify the targets and clutter points. Then selects the candidate tracks according to the logic-based algorithm. Finally, to eliminate false tracks, the correlation coefficient method is used to determine the track initiation results. Simulation results show that the algorithm has a lower false track probability compared to the conventional logic-based method in heavy clutter environment.

Due to the complexity and inaccuracy of the force model of low earth orbit(LEO) satellite, the ephemeris data is not accurate enough for translation compensation, which will cause the residual translation component error, then leading to the dissatisfaction of the range and azimuth resolution for imaging. To face the problem above, the secondary correction of translational component is carried out by updating the ephemeris data fitted by the ephemeris data calculated by orbit two lines elements(TLE) and center slant which is estimated with the real echo data of the space station through correlation method, after a coarse correction-the registration of ephemeris data and echo data-and then the Inverse synthetic aperture radar(ISAR) image of the space station is obtained by combining the polar format algorithm(PFA), providing the foundation for monitoring, identification and tracking of space targets. The imaging results prove the validity of the method, which can be referred to in the imaging of LEO satellite and the real data processing based on ephemeris data.

We propose a new algorithm, based on the graph generation method, to solve the problem of super edge-magic total labeling of graphs with a large number of vertices. Our main contributions to solving the graph labeling problem are: first, we generate the adjacent matrices of graphs composed of fans and stars. Second, we input these matrices to our proposed algorithm. Third, if the labeling is successful, the algorithm will output the corresponding super edge-magic total labeling matrices. Otherwise, no super edge-magic total labeling exists for the graphs involved. Forth, through the analysis of the results obtained by our algorithm, we conclude that regular labels are found in some of the graphs involved. Our algorithm can distinguish between graphs that have a super edge-magic total labeling and graphs that don't efficiently. We think that our proposed algorithm is innovative and can be adopted by other researchers.

A new algorithm for vertex-distinguishing edge coloring of graph is designed. First of all, by definition of vertex-distinguishing edge coloring, we know that there exist two constraints rules, one is the edge constraint and the other is the vertex color set constraint. Second, we established objective function according to the two constraint rules. Third, based on the exchange rules, an algorithm is designed to make the objective function meet the requirements. The algorithm steps are described in detail. The experimental results show that the vertex-distinguishing edge coloring obtained by the algorithm conforms to the coloring conditions. This algorithm can effectively solve this coloring problem of graphs with large number of vertices.

In recent years, the research and application of blockchain applied in e-government has been increasing. The characteristics of blockchain, such as non-tamper, verifiable and inspirable, have been well applied in confirming the rights, accountability, and assessment of government data resources, and other unresolved problems over the years. Well, the decentralized feature is naturally inconsistent with the strong central feature of the government so that the application of common blockchain technology in e-government cannot support the government regulatory mechanism effectively. To solve this problem, this paper proposes a strong supervision model of blockchain based on the Threshold Ring Signature algorithm (TRS). Further experiments show that the process and results of signature voting by using this model. Compared with the common blockchain, the strong supervision model can control the error rate of the chain information and increase the flexibility of blockchain so as to enhance the integration application of blockchain and e-government.

According to many experimental observations, neurons in cerebral cortex tend to operate in an asynchronous regime, firing independently of each other. In contrast, many other experimental observations reveal cortical population firing dynamics that are relatively coordinated and occasionally synchronous. These discrepant observations have naturally led to competing hypotheses. A commonly hypothesized explanation of asynchronous firing is that excitatory and inhibitory synaptic inputs are precisely correlated, nearly canceling each other, sometimes referred to as ‘balanced’ excitation and inhibition. On the other hand, the ‘criticality’ hypothesis posits an explanation of the more coordinated state that also requires a certain balance of excitatory and inhibitory interactions. Both hypotheses claim the same qualitative mechanism—properly balanced excitation and inhibition. Thus, a natural question arises: how are asynchronous population dynamics and critical dynamics related, how do they differ? Here we propose an answer to this question based on investigation of a simple, network-level computational model. We show that the strength of inhibitory synapses relative to excitatory synapses can be tuned from weak to strong to generate a family of models that spans a continuum from critical dynamics to asynchronous dynamics. Our results demonstrate that the coordinated dynamics of criticality and asynchronous dynamics can be generated by the same neural system if excitatory and inhibitory synapses are tuned appropriately., Author summary What is the dynamical state of cerebral cortex? Are neurons mostly uncorrelated, firing independently of each other? Or are neurons more coordinated, firing in sync with each other? The answers to these questions have fundamental implications for how the cortical neural population encodes and processes information. Here we show that two possible scenarios—criticality and asynchronous dynamics—that are typically considered incompatible can be attained in the same network by maintaining a certain kind of balance, while tuning the strength of inhibition relative to excitation along with the average synaptic strength.

Since Synthetic Aperture Radar (SAR) targets are full of coherent speckle noise, the traditional deep learning models are difficult to effectively extract key features of the targets and share high computational complexity. To solve the problem, an effective lightweight Convolutional Neural Network (CNN) model incorporating transfer learning is proposed for better handling SAR targets recognition tasks. In this work, firstly we propose the Atrous-Inception module, which combines both atrous convolution and inception module to obtain rich global receptive fields, while strictly controlling the parameter amount and realizing lightweight network architecture. Secondly, the transfer learning strategy is used to effectively transfer the prior knowledge of the optical, non-optical, hybrid optical and non-optical domains to the SAR target recognition tasks, thereby improving the model&rsquo, s recognition performance on small sample SAR target datasets. Finally, the model constructed in this paper is verified to be 97.97% on ten types of MSTAR datasets under standard operating conditions, reaching a mainstream target recognition rate. Meanwhile, the method presented in this paper shows strong robustness and generalization performance on a small number of randomly sampled SAR target datasets.

Data provenance technology and right confirmation technology have received great attention in recent years, because data exposure and abuse have been a serious problem with the rapid development of smart device and hospital. Based on the traceability and unchangeable properties of blockchain, we propose a model of personal health-related application data provenance and use smart contract to ensure right confirmation. We find that the user condition and use-right of personal health data can be effectively confirmed by using data provenance. In addition, the possible future extensions to personal health data on provenance are discussed.

Objective: Megavoltage computed tomography (MV-CT) is used for setup verification and adaptive radiotherapy in tomotherapy. However, its low contrast and high noise lead to poor image quality. This study aimed to develop a deep-learning-based method to generate synthetic kilovoltage CT (skV-CT) and then evaluate its ability to improve image quality and tumor segmentation. Approach: The planning kV-CT and MV-CT images of 270 patients with nasopharyngeal carcinoma (NPC) treated on an Accuray TomoHD system were used. An improved cycle-consistent adversarial network which used residual blocks as its generator was adopted to learn the mapping between MV-CT and kV-CT and then generate skV-CT from MV-CT. A Catphan 700 phantom and 30 patients with NPC were used to evaluate image quality. The quantitative indices included contrast-to-noise ratio (CNR), uniformity and signal-to-noise ratio (SNR) for the phantom and the structural similarity index measure (SSIM), mean absolute error (MAE), and peak signal-to-noise ratio (PSNR) for patients. Next, we trained three models for segmentation of the clinical target volume (CTV): MV-CT, skV-CT, and MV-CT combined with skV-CT. The segmentation accuracy was compared with indices of the dice similarity coefficient (DSC) and mean distance agreement (MDA). Main results: Compared with MV-CT, skV-CT showed significant improvement in CNR (184.0%), image uniformity (34.7%), and SNR (199.0%) in the phantom study and improved SSIM (1.7%), MAE (24.7%), and PSNR (7.5%) in the patient study. For CTV segmentation with only MV-CT, only skV-CT, and MV-CT combined with skV-CT, the DSCs were 0.75 ± 0.04, 0.78 ± 0.04, and 0.79 ± 0.03, respectively, and the MDAs (in mm) were 3.69 ± 0.81, 3.14 ± 0.80, and 2.90 ± 0.62, respectively. Significance: The proposed method improved the image quality of MV-CT and thus tumor segmentation in helical tomotherapy. The method potentially can benefit adaptive radiotherapy.

The spaceborne stripmap range sweep synthetic aperture radar (SS-RSSAR) is a new concept spaceborne SAR system that images the region of interest (ROI) with ROI-orientated strips, which, unlike the traditional spaceborne SAR, are allowed to be not parallel with the satellite orbit. The SS-RSSAR imaging is a challenging problem because echoes of a wide region have strong spatial varieties, especially in high-squint geometries, and are hard to be focused by a single swath. The traditional imaging algorithms could solve this problem by cost-ineffectively dividing an ROI into many subswaths for separate processing. In this paper, a new wide nonlinear chirp scaling (W-NLCS) algorithm is proposed to efficiently image the SS-RSSAR data in a single swath. Comparing with the traditional nonlinear chirp scaling algorithm, the W-NLCS algorithm is superior in three major aspects: the nonlinear bulk range migration compensation (RMC), the interpolation-based residual RMC, and the modified azimuth frequency perturbation. Specifically, the interpolation for the residual RMC, the most significant step in achieving the wide-swath imaging performance, is made innovatively in the time domain. The derivation of the W-NLCS algorithm, as well as the performance analyses of the W-NLCS algorithm in aspects of the azimuth resolution, accuracy, and complexity, are all provided. The presented approach is evaluated by the point target simulations.

This letter theoretically extends and applies the overlapped subaperture algorithm (OSA) for the challenging spotlight synthetic aperture radar (SAR) imaging in a quasi-forward-looking (QFL) geometry. A parameter-adjusting framework, in which the radar parameters relate closely to the geometry, is employed to mitigate the spatial signal coupling. By innovatively fitting the phase error induced by the planar wavefront assumption directly to the wavenumbers, both the linear and quadratic phase errors, which determine the geometrical distortion and the defocusing effects, respectively, can be accurately expressed and then corrected. When compared with the parameter-adjusting polar format algorithm proposed by the authors in previous work, the OSA is superior in contributing to an expanded imaging swath. The validity of the OSA methodology is demonstrated by the simulation of a Ka-band SAR working in the QFL diving mode with an extreme 85° squint angle.

This paper studies positioning issue in navigation using InSAR concept. A novel method based on wrapped interferometric phase registration, which differs from common InSAR based navigation method, is proposed. This method may be useful when navigating throw hills or areas with low SAR image characteristics but rich terrain elevation features. Accurate DEM data is needed to generate ideal interferometric phase template for registration. Three dimensional positioning is achieved by estimating vertical bias and horizontal bias in order. The proposed method works well under different squint angle. Simulations are carried out to verify the method. Precision of the method under different squint angle is also exhibited in simulations.

The spatial localization of interference source is a critical procedure in Radio frequency interference (RFI) suppression. It is not suitable to use traditional multi-station interference localization methods in most synthetic aperture radar (SAR) systems. This paper proposes a novel method of single RFI localization based on conjugate cross-correlation of dual-channel SAR signals. First, interference detection technology is applied to check whether echo signal is interfered. Then, conjugate cross-correlation is performed on SAR signals, and the distance difference between interference source and receiving channels is calculated. Finally, spatial localization equations of interference source are established, and optimization algorithm is applied to solve the coordinates of interference source. The results of simulation experiments illustrate the effectiveness of the proposed method.

The classical linear frequency modulated waveform (LFMW) for synthetic aperture radar (SAR) can achieve a high range resolution and a superior signal-to-noise ratio (SNR) for point targets. Therefore, numerous researches got increasingly passionate about increasing resolution and SNR, rather than information acquisition of LFMW. This paper quantitatively analyzes and evaluates target information acquisition ability of LFMW, which is conducive to target perception. Firstly, a factor contributing to information acquisition of LFMW, is obtained under the discretized signal model. Secondly, a connection between LFMW and eigenvectors of target feature matrix is derived through the discrete Fourier bases, thus the information acquisition of LFWM is presented. Finally, the numerical simulations are made for the stochastic extended targets information acquisition of LFMW, and the results are employed to verify the theory analysis.

According to many experimental observations, neurons in cerebral cortex tend to operate in an asynchronous regime, firing independently of each other. In contrast, many other experimental observations reveal cortical population firing dynamics that are relatively coordinated and occasionally synchronous. These discrepant observations have naturally led to a lively debate surrounding discrepant hypotheses. A commonly hypothesized explanation of asynchronous firing is that excitatory and inhibitory neurons are precisely correlated, nearly canceling each other, resulting in the so-called ‘chaotic balanced state’. On the other hand, the ‘criticality’ hypothesis posits an explanation of the more coordinated state that also requires a certain balance of excitatory and inhibitory interactions. Both hypotheses claim the same qualitative mechanism - properly balanced excitation and inhibition. Thus, a natural question arises: how are the chaotic balanced state and criticality related, how do they differ? Here we propose an answer to this question based on investigation of a simple, network-level computational model. We show that the strength of inhibitory synapses relative to excitatory synapses can be tuned from weak to strong to generate a family of models that spans a continuum from ‘criticality’ to the ‘chaotic balanced state’. Our results bridge two long-standing competing hypotheses and offer a possible explanation of discrepant experimental observations: neuromodulatory mechanisms that tune the strength of excitatory and inhibitory synapses may tune cortical state from criticality to the balanced state, and to intermediate states between these extremes.Author summaryWhat is the dynamical state of cerebral cortex? Are neurons mostly uncorrelated, firing asynchronously with each other? Are synchronous oscillations important? The answers to these questions have fundamental implications for how the cortical neural population encodes and processes information. Here we show that two possible scenarios - criticality and the chaotic balanced state - that are typically considered incompatible can be attained in the same network by maintaining a certain kind of balance, while tuning the strength of inhibition relative to excitation.

In this paper, a novel method is proposed to solve the problem of insufficient representation of single character-level features or word-level features. In view of the short length, sparseness and strong context dependencies of short text, our method takes word-level vectors and character-level vectors as inputs simultaneously, and encodes sentence semantics by two Long Short-Term Memory (LSTMs) or bidirectional Long Short-Term Memory (BiLSTMs). The outputs of the entire sentence combined two outputs from word-level vectors and character-level vectors. For Chinese short text classification, our experiments show that the combination of word embedding and character embedding can complement each other in the sentence semantic representation, which helps to improve the classification performance of Chinese short text.

With the development of mobile communication and Internet technology, the scale of travel time and space data has expanded dramatically, resulting in massive POI data. How to organize POI data efficiently and completely is extremely urgent. The conventional POI data organization methods are generally mixed organizations, and there is no special organization for specific tourism information, which leads to the irregular management of tourism information and restricts the popularized service of tourism information. In this paper, the POI data is organized by the AP-MRO method for tourism data, and the analytic hierarchy process is used to analyze the tourism decision-making, and the best tourism plan is obtained through calculation.

Synthetic aperture radar (SAR) systems are susceptible to radio frequency interference (RFI). The existence of RFI will cause serious degradation of SAR image quality and a huge risk of target misjudgment, which makes the research on RFI suppression methods receive widespread attention. Since the location of the RFI source is one of the most vital information for achieving RFI spatial filtering, this paper presents a novel location method of multiple independent RFI sources based on direction-of-arrival (DOA) estimation and the non-convex optimization algorithm. It deploys an L-shaped multi-channel array on the SAR system to receive echo signals, and utilizes the two-dimensional estimating signal parameter via rotational invariance techniques (2D-ESPRIT) algorithm to estimate the positional relationship between the RFI source and the SAR system, ultimately combines the DOA estimation results of multiple azimuth time to calculate the geographic location of RFI sources through the particle swarm optimization (PSO) algorithm. Results on simulation experiments prove the effectiveness of the proposed method.

The data sharing of electronic medical records (EMRs) has great positive significance for research on disease and epidemic prevention. Recently, blockchain-based EMRs storing and sharing schemes have brought many benefits. Nevertheless, the existing medical data sharing solutions cannot fully meet the query requirements in multi-user scenarios where data users with different identities have different needs for data usage. In this paper, we propose a block construction method for storing EMRs. We combine the characteristics of the Merkle tree and prefix dictionary tree in the block to reduce the time consumption of queries in the block. Besides, the block header is identified by four attributes, namely hospital ID, Department ID, disease type, and time period. According to the query conditions, the block where the requested data is located is quickly found. In the blockchain-based data sharing model of EMRs, data requesters with different identities can make combined queries with multiple conditions according to their own needs, and then the block and transactions in the block are located accurately. In conclusion, we are the first to build the block based on disease types and our model supports more flexible combined queries with multiple conditions.

Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR), most algorithms of which have employed and relied on sufficient training samples to receive a strong discriminative classification model, has remained a challenging task in recent years, among which the challenge of SAR data acquisition and further insight into the intuitive features of SAR images are the main concerns. In this paper, a deep transferred multi-level feature fusion attention network with dual optimized loss, called a multi-level feature attention Synthetic Aperture Radar network (MFFA-SARNET), is proposed to settle the problem of small samples in SAR ATR tasks. Firstly, a multi-level feature attention (MFFA) network is established to learn more discriminative features from SAR images with a fusion method, followed by alleviating the impact of background features on images with the following attention module that focuses more on the target features. Secondly, a novel dual optimized loss is incorporated to further optimize the classification network, which enhances the robust and discriminative learning power of features. Thirdly, transfer learning is utilized to validate the variances and small-sample classification tasks. Extensive experiments conducted on a public database with three different configurations consistently demonstrate the effectiveness of our proposed network, and the significant improvements yielded to surpass those of the state-of-the-art methods under small-sample conditions.

In this paper, firstly we model the channel impulse characteristic by combining the Monte Carlo method with Fournier-Forand (FF) phase function. Then we analyze the effects of various field of views (FOVs) and transmission distances on the channel impulse response and found that temporal dispersion has rigorous bandwidth limitations on the undersea wireless optical communication (UWOC) system in the harbor water. Furthermore, we propose to adopt the maximum likelihood sequence estimation (MLSE) method at the receiver to mitigate the bandwidth limitations and demonstrate that MLSE has great potential to improve system performance.

For cutting-edge protected agriculture, it is a leading trend to build an unmanned plant factory that requires robotic machinery as basic autonomous production units. However, the scheduling and cooperation among agricultural machines are rarely investigated. This paper reports on a modular logistics system based on the motion decomposition, and the objective is to develop a fully interrelating intelligent agricultural production system. Firstly, an unmanned ground vehicle is developed for transferring the intelligent agricultural machines such as agricultural robots. In consideration of the complex terrain features, both automatic mode and manual mode are available. The multifunctional smart remote-control mode is designed for users to control the logistics system accurately and the application of tablet PC provides an excellent user experience. Secondly, a control and mechanical system is designed to dock and transfer a harvesting robot. The safety and reliability of this system are jointly ensured by its movement pattern, the design of guide rails on the slope, and the precise positioning. Taking the performance on the test field into consideration, because the performance showed efficiency and accuracy of the system, future plans are highlighted to indicate the technical trends of unmanned plant factory.

In this paper, the actual project of cable-stayed bridge with single pylon is taken as the calculation object. This bridge has main span of 2 × 132m, and it is tower-girder consolidation system. The finite element model is established using Midas/civil software to calculate and analyze the bridge's behavior in construction and service status. The calculation includes the force calculation of the main girder, the central pylon and cables; also includes cracking resistance and stiffness checking computation of main girder in the serviceability limit state, the stress checking computation of the main girder, the central pylon and stay-cables also in the serviceability limit state. Calculations results prove that the bridge's static force performance meets the specification requirements both in the construction and service status. So it can provide a reference for the design scheme selection and calculation analysis of similar bridges.