Abstract:Extreme marine weather phenomena have an important impact on the coastal area. Researchers have made great progress in marine extreme meteorology prediction with the help of marine big data and deep learning algorithms. In this paper， taking typical multi-scale marine extreme weather phenomena—El Ni?o， typhoon， short-term precipitation as examples， we firstly introduce the mainstream marine extreme meteorology forecast algorithms in recent years， which are mainly divided into numerical model-based methods and artificial intelligence-based algorithms. Then，we analyze the challenges and opportunities of marine extreme meteorology prediction， and summarize the research advances of various methods in detail. And， we discuss the advantages and disadvantages of existing algorithms through experiments. Finally， we briefly look forward to the development direction of marine intelligent meteorology prediction based on marine big data.

Abstract:Multi-source data is a complex data type that integrates multiple information sources or data sets. Its main feature is that different information sources imply different knowledge structures， and represent and describe samples and relationships between samples from different perspectives. How to fuse and integrate multi-source data cooperatively and how to quickly mine the overall decision-making knowledge for users from different viewpoints have become a scientific problem that needs to be solved urgently in the field of data science. Classical rough set theory， multi-granularity method， evidence theory and information entropy are common and effective multi-source information fusion methods， which have been widely concerned and achieved fruitful results. Therefore， this paper summarizes the work of multi-source information fusion based on granular computing， reviews the basic concepts and main research ideas of each information fusion method， and puts forward some problems in the field of multi-source information fusion. The obtained results can provide a theoretical reference for the follow-up research in this field.

Abstract:Aiming at the problem that the current mainstream dual-branch single-channel speech enhancement methods only pay attention to the full frequency band information while ignoring the subband information， an interactive dual-branch model based on the critical frequency band of the human ear is proposed. The main method is to implement the division method of simulating the critical frequency band of the human ear on the complex spectrum branch to process the signal in frequency division and extract sub-band information. The whole frequency band of the signal is directly processed on the amplitude compensation branch， and the information of the whole frequency band is extracted. The complex spectrum branch is responsible for initially recovering the amplitude and phase of the clean speech signal. At the same time， the subband intermediate features learned by the branch are transferred to the amplitude compensation branch by specific modules for compensation. The output on the amplitude compensation branch will further compensate the amplitude of the output on the complex spectrum branch to achieve the purpose of recovering the clean speech spectrum. Experimental results show that the proposed model is superior to other advanced models in restoring speech quality and intelligibility.

Abstract:Deep neural network based full-band speech enhancement systems face challenges of high demand of computational resources and imbalanced frequency distribution. In this paper， a light-weight full-band model is proposed based on dual path convolutional recurrent network with two dedicated strategies， i.e.， a learnable spectral compression mapping for more effective high-band spectral information compression， and the utilization of the multi-head attention mechanism for more effective modeling of the global spectral pattern. Experiments validate the efficacy of the proposed strategies and show that the proposed model achieves competitive performance with only 0.89×10⁶ parameters.

Abstract:Considering that the spatial relationship between channels affects the noise reduction， graph signal processing can capture the potential relationship. If the spatial physical distribution map is directly used， its time-varying characteristics cannot be reflected in real time. Therefore， we propose a multi-channel speech enhancement method based on joint graph learning. Firstly， we propose a joint time-space graph learning method， which jointly optimizes the array space graph and the speech frame inner graph， for the sake of minimizing the sum of the smoothness of the multi-channel noisy speech signal on the spatial graph， the smoothness of the nosiy speech signal from the reference channel on the speech frame graph， the sparsity of the Laplace matrix and the sparsity of the adjacency matrix. Based on the learned space graph and frame inner graph， the time-space joint graph of multi-channel speech signal is constructed. On this basis， the multi-channel speech graph signal is enhanced by applying the joint graph transform and the fixed beam forming （FBF） method. Experimental results show that the proposed joint graph learning based FBF （JGL-FBF） method can significantly improve the signal-to-noise ratio （SNR） of enhanced speech and perceptual evaluation of speech quality （PESQ） compared with the traditional FBF method. In addition， the experimental results also show that the accuracy of delay compensation affects the speech enhancement performance of JGL-FBF.

Abstract:In the complex application environment of the global satellite navigation system （GNSS）， where various kinds of suppressive interference and spoofing randomly exist， the traditional interference suppressing method that first estimates the interference parameters and then canceles interference signal， will be designed difficultly and has low generality， because the special parameter estimators and the interference reducing methods are needed for various types of interference. Therefore， an interference suppression scheme based on deep residual neural networks （DRNNs） is proposed in this paper. First， the corresponding DRNN is built and trained for each typical GNSS interference. It can directly extract the target satellite signal from the interfered signal. Second， according to the interference classification and recognition result， the corresponding DRNN is selected. The time-frequency two dimensional （2D） signals obtained by short-time Fourier transform over the received one-dimensional signal are then entered into the chosen DRNN. The output is the 2D time-frequency spectrum of the useful signal， where the impact of the interference has been suppressed. In our scheme， the same procedure is applied for different kinds of suppressive interference and spoofing. It is not required to design the special designs about the parameter estimation and the interference reduction for various interferences. Experimental results show that the proposed scheme can effectively suppress various GNSS interference， compared with the traditional scheme. It demonstrates a certain of commonality.

Abstract:In order to solve the space-time discontinuity problem of unmanned aerial vehicle（UAV） non-stationary channel parameters， this paper proposes a non-stationary geometric stochastic model of UAV that supports three-dimensional motion and posture rotation， which is used to describe and simulate the real characteristics of UAV multiple-input multiple-output communication channel. Based on the concept of “accurate qualitative” in QuaDRiGa， the model update time-varying channel parameters based on the topological relation of receiver and transmitter， improve the channel power calculation method considering the probability of path birth and death， and introduce posture phase matrix to describe UAV posture rotation， so as to realize smooth evolution of UAV non-stationary channel parameters. The numerical simulation results show that the model proposed in this paper can ensure the smooth evolution of parameters such as power and angle， and the autocorrelation function of the output channel is non-stationary and significantly affected by UAV posture. The proposed model can be used in UAV communication system design and algorithm optimization.

Abstract:Data distribution service（DDS） is a reliable real-time data communication middleware standard. It is oriented to a distributed environment based on the publish/subscribe model. It has been widely used in various fields. However， there are few achievements in existing research involving DDS security technology. There are many security threats to the publishing and subscribing system in practice. In order to establish a flexible and reliable security mechanism to ensure the security of publishing and subscribing information， a data-centric access control scheme is proposed. On the basis of attribute encryption， the access tree structure is optimized， and the attribute trust mechanism is added in combination with the publishing and subscribing environment. Afterwards， the publicating and subscripting information is encrypted and matched by formulating attribute connection and authorization strategies， and a DDS access control model is established to control the interaction of information in the publicating and subscripting system and realize the safe distribution of data. The experimental verification shows the solution can deal with several security threats in DDS， guarantee the confidentiality of publishing and subscribing information， as well as realize the system’s access control to specific information， and publishers and subscribers do not need to share keys， reducing the overhead of key management.

Abstract:Multi-source domain generalization （DG） aims to train a model uses semantic information of different domains and can be generalized to unknown domains. In the medical image， the gap between different domains is relatively large， and the model will suffer from performance drop in the unknown domain. In order to solve this problem， this paper proposes a network structure which encodes images for features and decodes domain specific features. The model uses a generic encoder， which learns all source domains for the domain-invariant features， and several domain-specific decoders to reconstruct the original images to promote the ability of extracting image features. Meanwhile， these decoders also help to generate transferred image to engage in adversarial learning with images of source domains in order to improve the encoder’s ability of learning invariant features. In addition， we also introduce a special Cutmix strategy which change foreground information of different domain images to augment the data set in the model to enhance the generalization ability of the model and further improve the performance of our network structure. In two medical image segmentation tasks， a large number of experimental data show that the proposed model has excellent performance compared with the existing advanced models. In addition， a series of ablation experiments are carried out to prove the effectiveness of the model.

Abstract:Effective acquisition of spatial structural features of point cloud data is the key to semantic segmentation of point clouds. To solve the problem that the previous methods do not make good use of global and local features， a new spatial structure feature， point box feature， is proposed for semantic segmentation. A network framework of encoding-decoding structure is designed. The global spatial and local neighborhood features of point clouds are learned by using the geometric structure feature module during the downsampling process， and the full size feature map is restored step by step in the upper sampling process for semantic segmentation. The geometric structure features module contains two sub-modules， one is the global features module， which learns the “box” features of points to represent the rough geometric features of point clouds in the sampling space. Another is the local features module， which uses feature extraction， the attention mechanism structure， to represent precise， fine-grained geometric characteristics of point clouds within local neighborhoods. Experiments are performed on the public dataset S3DIS and Semantic3D and compared with other methods. The results show that mIoU is ahead of most of the current mainstream methods， and some of the detail class IoU is the highest.

Abstract:Low resolution is an important factor that affects the accuracy of face recognition. To overcome the limitation of low-resolution facial images on face recognition， one effective solution is adopting super-resolution methods to reconstruct low-resolution images and then identify the generated facial images. However， existing super-resolution methods typically fail to consider facial identity preservation during reconstruction， which directly results in poor face recognition performance of reconstructed images. To address the issue mentioned above， this paper proposes a face super-resolution reconstruction method with identity preserving， called IPNet. This method can simultaneously improve the quality of low-resolution facial images and preserve the identity of reconstructed images. IPNet consists of a semantic segmentation network and a face generator. The semantic segmentation network is introduced to extract low-dimensional latent code and multi-resolution spatial features. Then， the extracted features guide the face generator to output super-resolution images similar to the authentic images. Furthermore， we introduce the face recognition network to integrate the face identity information into the super-resolution model， thus maintaining the identity of reconstructed facial images consistent with original images. Experimental results show that IPNet achieves better results than other comparison methods in terms of both super-resolution image quality and identity preservation， demonstrating effectiveness of the proposed method.

Abstract:Aiming at the problem that the fusion method used by the existing multi-scale object detection model in the face of scale variation and occlusion scene is not sufficient， and does not capture the long-distance dependency relationship， channel feature fusion aggregation module and non-local feature interaction module are designed to learn the correlation between different channel features and capture the long-distance dependence between feature maps. In addition， the current detection architecture is based on single pyramid detection structure， which exists information loss. In this paper， a double pyramid structure is designed， and the proposed fusion method is combined with the double feature pyramid structure to supplement the fusion feature information on the basis of preserving the original feature information. Experimental results on public datasets KITTI and PASCAL VOC show that the proposed method has higher detection accuracy than other advanced work， proving its effectiveness in object detection task.

Abstract:In order to solve the problem of low accuracy of Transformer encoder caused by the loss of person image blocks information and insufficient expression of person local features in person re-identification， an improved Transformer encoder and feature fusion algorithm for person re-identification is proposed. This algorithm uses relative position encoding to solve the problem that Transformer will lose the relative position information of person image blocks during attention operation so that the network can focus on the semantic feature information of person image blocks， thus enhancing the ability to extract pedestrian features. Secondly， the local patch attention module is embedded into the Transformer network to weighted strengthen the local key feature information and highlight the significant features of the person area. Finally， the fusion of global and local information features is used to achieve complementary advantages between features and improve the recognition ability of the model. In the training stage， Softmax and triple loss functions are used to jointly optimize the network. The proposed algorithm is experimentally compared and analyzed on the mainstream datasets of Market1501 and DukeMTMC-reID. The Rank-1 accuracy reaches 97.5% and 93.5% respectively， and the mean average precision （mAP） reaches 92.3% and 83.1% respectively. The experimental results show that the improved Transformer encoder and feature fusion algorithm can effectively improve the accuracy of person re-identification.

Abstract:Shapelet is the most identifiable subsequence in time series， which has been extensively studied by researchers from various fields since it was proposed. In this process， many effective shapelet discovery techniques have been proposed for time series classification. However， candidate shapelets of multivariate time series may have different lengths and different sources of variables， making it difficult to directly compare them， which presents a unique challenge to the classification method of multivariable time series based on shapelet. we propose Multi-shapelet， a multivariate time series classification method based on unsupervised representation learning and shapelets. Firstly， Multi-shapelet uses a hybrid model DC-GNN （Dilated convolution neural network and graph neural network） as an encoder to embed candidate shapelets of different lengths into a unified shapelet selection space for comparison between shapelets. Secondly， a new loss function is proposed to train the encoder in an unsupervised learning manner， so that after DC-GNN encodes the shapelet to obtain the corresponding embedding， the topology and the original space formed by the relative positions between the embeddings corresponding to the shapelet belonging to the same class. The relationship between the topologies formed by the relative positions of the shapelet in the middle is closer to a proportional reduction， which is very important for the subsequent similarity-based pruning process. Finally， the K-means clustering and simulated annealing algorithm are proposed to prune and select shapelets to select a set of shapelets with strong classification ability. Experimental results on 18 UEA multivariable time series datasets show that the overall accuracy of Multi-shapelet is significantly better than other methods.

Abstract:Microorganisms have a direct impact on human health， and the analysis of relevant data is helpful for disease diagnosis. However， the collected data suffers from two problems： class imbalance and high sparseness. Existing oversampling methods can alleviate the class imbalance of data to a certain extent， but it is difficult to cope with the high sparsity of microbial data. This paper proposes a data augmentation algorithm that fuses matrix factorization and cost-sensitive， which consists of three techniques. First， the original matrix is decomposed into a sample subspace and a feature subspace. Second， the positive vectors of the sample subspace and their neighbor vectors are used to generate synthetic vectors. Finally， the synthetic vectors are filtered according to their distance from all negative vectors. The proposed algorithm is compared with five oversampling algorithms on 8 microbial datasets. The results show that the proposed algorithm can enhance the diversity of positive samples and identify more positive samples with lower classification cost.

Abstract:In the era of big data， the potential value of data makes it one of the important assets. The illegal tampering， correction and illegal distribution of data bring great challenges to tracing the source of data leakage. Digital fingerprint technology can be applied in the field of traceability of data leakage， that is， a sequence of unique identification of user information is embedded in the data. Multiple users conspire to attack data and leak data， thereby destroying the fingerprint information embedded in it to escape accountability. Anti-collusion coding can solve this problem. Aiming at the problems that the existing digital fingerprint encoding cannot meet the data redistribution requirements and the digital fingerprint embedding causes large data distortion， this paper uses the balanced incomplete block design （BIBD） as the outer code and the C code after codeword expansion as the inner code to construct a redistribution anti-collusion fingerprint coding （RD-ACC）. On this basis， a database fingerprint algorithm based on multi-objective optimization is proposed to ensure high robustness of digital fingerprints under the condition of small database distortion. The extracted RD-ACC can effectively resist intra-group and inter-group multi-user collusion attack. Experimental results show that the algorithm can realize the data redistribution operation with less data distortion， and resist the collusion attack to trace the source of leaks.

Abstract:Density-based spatial clustering of applications with noise （DBSCAN） can find clusters of different densities and sizes， is also robust to noise， and is widely used in data mining tasks. DBSCAN needs to adjust the parameters MinPts and Eps to achieve a better clustering effect， but it often affects the performance of DBSCAN in the process of searching for the optimal parameters. This article optimizes DBSCAN from two aspects. On one hand， a parameter-free method is proposed to optimize DBSCAN global parameter selection. The parameter-free method uses the natural nearest neighbor to obtain the natural feature value of the data set， and uses the natural feature value as MinPts. Then， the natural feature set is calculated according to the natural feature value， and three strategies （i.e. statistics of minimum， mean and maximum） are used to obtain the Eps values by using the data distribution characteristics of the natural feature set. On the other hand， it uses the graphics processing unit （GPU） of the real-time acceleration platform for integrated data science （RAPIDS） platform to accelerate the convergence of DBSCAN algorithm. The experimental results show that the proposed method can optimize DBSCAN parameter selection while obtaining the comparable clustering results of density peaks clustering （DPC） algorithm.

Abstract:Concept cognition and knowledge discovery from network data are important research directions of machine learning and artificial intelligence under the network background， and have been introduced into the study of recommendation system. The existing recommendation methods based on concept lattice ignore the network structure relationship between nodes. At the same time， the efficiency of constructing concept lattice is low and the constraints of constructing concept set are strict， which is difficult to realize in large-scale social networks. In order to solve these problems， this paper integrates the topology of complex networks and weaken-concept similarity under the framework of network formal context， and proposes a group recommendation algorithm based on weaken-concept similarity. Firstly， the importance of attributes is described by defining attribute degree and attribute density， and then the expert nodes are determined by using the improved node influence. Secondly， the community is divided by expert nodes， the group recommendation research is carried out by using the lower limit similarity of attribute weaken-concept in the divided community， and then the recommendation rules are obtained and the group recommendation is applied to the corresponding communities. Finally， the influence of various parameters on the algorithm is analyzed on MovieLens and Filmtrust datasets， and reasonable values of the parameters are determined. After that， the proposed algorithm is compared with other recommended algorithms， and the experiments show that the proposed algorithm is effective.

Abstract:To solve the problems of the traditional A* algorithm and the traditional dynamic window approach （DWA） in mobile robot path planning， a dynamic path planning method combining the improved A* algorithm and the improved DWA is proposed. First， the 16-neighborhood and 16-direction path search method is adopted to expand the path search field and reduce the number of the nodes accessed and the turning angles. Second， the heuristic function is optimized to enhance the purpose of the path search. Next， the redundant point deletion strategy is adopted to reduce the number of the turning points and further improve the smoothness of the path. Third， the path corner is processed by the B-spline curve， and the path is relatively smooth. Then， the sensitivity of obstacle avoidance can be improved by classifying and treating obstacles differently and adding the speed adaptive factor in the evaluation function of DWA. Finally， through three parts of the simulation experiments with the other algorithms， and the priority strategy simulation experiments， the effectiveness of the improved A* algorithm and the superiority of the fusion method in obstacle avoidance are verified.

Abstract:In order to realize precise location for microwave wideband signals with high accuracy， this paper investigates the correlative interferometer-based direction finding algorithm with uniform circular array and microwave photonic phase detector. Compared with the low operating frequency and narrow bandwidth of the electronic microwave phase measurement， a photonic microwave phase measurement using dual polarization Mach-Zehnder modulator （DPol-MZM） is presented with the benefits of high frequency band， wide bandwidth， ultralow loss and immunity to electromagnetic interferences. The output optical power is related to the phase shift which can be employed for the phase shift measurement. Owing to the accuracy monitoring， the proposed phase detector can measure the phase shift from -180°to 180°with ±2°measurement error. In addition， the cosine function has been chosen to replace the conventional correlation function to solve the phase ambiguity. Meanwhile， an appropriate search algorithm with large angle step should be selected to ensure that the direction finding algorithm can be operated in real time. Besides， the conicoid interpolation fitting has been exploited for compensation the measurement error caused by the large search step. Finally， representative simulations have been presented to demonstrate the validity of the microwave photonic phase measurement and the direction finding algorithm.