Abstract:Low-altitude intelligent network is an instrumental infrastructure for the outgrowth of low-altitude economy. However， the safety control of the unmanned aerial vehicles （UAVs） presented in such complex system faces multiple security challenges such as airspace security， data security， and spectrum security. To address these three issues simultaneously， a blockchain-based three-sided collaborative regulatory architecture， with the use of both “on-chain” and “off-chain” information， is proposed. The “on-chain” contains identity and registration information of UAVs， while the “off-chain” contains automatic dependent surveillance-broadcast （ADS-B） information and spectrum information. To solve the problem of cross-domain authentication， an effective signature algorithm is developed， which is based on the ADS-B information and certificateless signature. Furthermore， due to the lack of error correction mechanism in the ADS-B protocol， errors are easily incurred by channel noise and interference during the transmission of the ADS-B information. Consequently， the hash verification may fail. In order to alleviate such signature failure， a cross-layer signature algorithm based on error correction code is designed for correcting errors. The proposed blockchain-based three-sided collaborative regulatory platform has been well experimented over the Yangtze River low-altitude demonstration pilot zone and achieved great success.

Abstract:To address the issue of resource allocation for multiple electronic jamming unmanned aerial vehicles （UAVs） against multiple multifunctional radars in the low-altitude intelligent network cooperative cognitive jamming decision-making process， a cognitive jamming decision-making approach based on digital twinning and deep reinforcement learning is proposed. Firstly， a cognitive jamming decision-making system model is established by treating the cooperative electronic jamming problem as a Markov decision process. Considering the constraints related to jamming target， jamming power， and jamming pattern selection comprehensively， the agents’ action space， state space， and reward function are constructed. Secondly， an adaptive learning rate proximal policy optimization （APPO） algorithm is proposed based on the proximal policy optimization （PPO） algorithm. Additionally， to enhance the training speed of the deep reinforcement learning algorithm in a high-fidelity manner， a digital twin-based cooperative electronic jamming decision-making model training method is presented. Simulation results demonstrate that compared with existing deep reinforcement learning algorithms， the interference efficiency of the APPO algorithm is improved by more than 30%， and the proposed training method increases the model training speed by more than 50%.

Abstract:Because of the small space range，slow target speed， and mixed environmental elements of low-altitude flight， the traditional latitude and longitude characterization cannot meet the requirements of low-altitude fine management in the Smartlink environment. Therefore the digital low-altitude airspace raster characterization metrics and optimal calibration problems are studied. Firstly， the quantitative characterization rules of multi-dimensional low-altitude airspace structural elements are constructed from the perspective of “point-line-plane”， the quantitative characterization method of multi-level raster in low-altitude airspace is proposed. Then， by determining the “point-line-plane” positional relationships of different airspace rasters， we propose a topological relationship metric of low-altitude airspace based on the raster intersection matrix. Finally， considering the optimization objectives of low-altitude unmanned aerial vehicle（UAV） collision index and low-altitude raster utilization index， as well as the node-raster matching constraints， spatial position constraints， and safety constraints of UAVs and UAVs/obstacles， we establish a multi-dimensional performance oriented optimal calibration model for the raster granularity of the low airspace， and evaluate the effectiveness and efficiency of the proposed method for the typical mission scenarios of the low airspace. The validity and optimization effect of the proposed method are verified and analyzed for typical urban low altitude flight scenarios. The experimental results show that the proposed method can optimally configure the digital low altitude raster granularity for any low-altitude airspace and UAV mission with acceptable UAV collision index and raster utilization index， so as to realize the safety and high efficiency of low altitude flight activities. The research results have certain theoretical value and application significance to support the fine management of digital low altitude airspace and the fusion operation of heterogeneous aircraft.

Abstract:As an important component of low-altitude intelligent networking， unmanned aerial vehicles （UAVs） have been widely used in the field of wireless communications. Nevertheless， the existing solutions often encounter numerous challenges when dealing with the continuously evolving scale and topology of UAV networks， such as slow convergence speed， insufficient real-time response capability， high training costs， and limited generalization abilities. To address these issues， this paper proposes an observation representation and decision-making scheme based on graph neural networks （GNNs） for UAV networks. The study initially models the relationships between UAVs and their observational entities using graph modeling techniques， designs a GNN-based representation scheme， and utilizes machine learning algorithms for pre-training to adapt to the dynamically changing observation space. For the dynamic characteristics of the decision space， the paper further introduces an edge-decision-based GNN model， which enhances adaptability to the dynamic decision space through graph modeling and edge weight fitting. Moreover， through the study of two UAV network cases， the effectiveness and superiority of the proposed scheme are validated， demonstrating its potential in practical UAV network applications.

Abstract:With the development of networked and intelligent unmanned aerial vehicles （UAVs）， they have gradually become an important component of the low-altitude intelligent network （LAIN）. However， the effective management of UAV platforms in the LAIN still faces severe challenges. Based on the subtle features of UAV signals， individual identification of UAVs can be achieved， and illegal UAVs can be detected， thereby realizing the identification and management of UAVs in the LAIN. In response to the problem of complex channel environments and the inability to obtain illegal UAV signal samples in advance in the low-altitude domain， this paper proposes an open set identification method for UAVs based on differential time-frequency and multi-center OpenMax. Firstly， this paper proposes channel-independent differential time-frequency features to reduce the impact of multipath channel environments on radio frequency fingerprinting （RFF） features and uses data augmentation to improve the accuracy and robustness of the identification model. Secondly， this paper uses multi-center OpenMax to replace the neural network’s SoftMax layer for open set identification of UAVs. Finally， the loss function of the neural network is improved to increase the accuracy of open set recognition. The proposed algorithm is validated using real-world data. When the openness is 0.087， the open set recognition accuracy reaches 93.23%， an increase of 7.61% and 13.4% compared with the benchmark algorithms. The algorithm proposed in this paper can effectively identify individual UAVs and detect illegal UAVs appearing for the first time in complex channel environments.

Abstract:The prosperity of the low-altitude ecology has continuously promoted the transformation of intelligent network services from the terrestrial to the low-altitude airspace. Low-altitude services and applications have become large-scale， collaborative， and intelligent. These trends have put forward extreme requirements for cross-domain collaboration capabilities， processing efficiency， and the security and trustworthiness of data and operations. Low-altitude intelligent network cluster collaboration using multi-device joint computing can improve the processing efficiency of complex and large-scale tasks in low-altitude intelligent networks. However， the existing schemes still have problems such as lack of cross-domain collaboration， deficiency in security and trustworthiness， and insufficient flexibility in centralized resource scheduling. Blockchain has the characteristics of immutability， openness， transparency， and collective maintenance， which is suitable for establishing efficient collaborative trust. This paper proposes a blockchain-enhanced trustworthy collaboration architecture for low-altitude intelligent network to provide on-chain cross-domain collaborative computing and trusted status synchronization services among heterogeneous low-altitude intelligent devices. We also design a multi-level consensus protocol to ensure the security and trustworthiness during the collaborative computation process. And we further analyze the freshness of the on-chain status， and propose an on-chain state correction algorithm and an efficient cluster-forming strategy for low-altitude intelligent nodes based on a queueing model. The simulation results show that the proposed architecture and protocol can improve the overall performance in terms of collaborative processing efficiency and network resource utilization.

Abstract:Aiming at the problem of information freshness in unmanned aerial vehicle （UAV） assisted content distribution system， a UAV caching and trajectory optimization algorithm based on age of information （AoI） is proposed to alleviate the problem of long time unanswered user requests in hotspot areas. The problem of minimizing the average cost of accessing the requested content for all users is established by optimizing the ground user clustering， the UAV caching policy and the trajectory within the limited cache capacity and coverage of the UAV. The radius of coverage of UAVs is used as the radius of clustering， and the affinity propagation （AP） clustering algorithm is used to cluster the ground users. The UAV caching problem in this paper is transformed into the 01 backpacking problem， which is solved using the dynamic programming （DP） algorithm. UAV flight trajectories are solved by the genetic algorithm （GA）. Simulation results show that the algorithm proposed in this paper can effectively reduce the average cost for users to obtain the requested content.

Abstract:To address the complex handover problem in multi-layer heterogeneous integrated networks （MLHetINet）， this paper develops a staring beam and interference cancellation algorithm to get valid data for handover analysis， simplifies the beam alignment and acquisition process， expands the coverage of the air-based network， and reduces the complexity of cross-layer handover. Firstly， aiming at the relatively high-speed movement between the ground terminal and the air base station， this paper develops a dynamic staring multi-beam forming algorithm to adaptively adjust the antenna phase and weight， and generates the main lobe in the target direction and the zero-trap for the interference source to achieve airspace isolation， thus simplifying the complexity of handover analysis. Then， considering the complexity of the air-ground channel， a multi-order interference cancellation algorithm based on column norm grouping sorting is proposed to further improve the detection accuracy of the target signal and the accuracy of handover analysis. Finally， based on the staring beam and interference cancellation algorithm， an independent handover protocol is designed for handover events in the space-air-ground three-dimensional MLHetINet， which significantly reduces the consumption of network resources. Through simulation， it is verified that the user information rate in the three-dimensional MLHetINet is significantly improved compared with that in the traditional ground network and space-ground integrated network.

Abstract:Deep learning techniques have greatly improved the classification accuracy of synthetic aperture radar (SAR） images target， but the security of SAR image classification systems is threatened by the inherent vulnerability of neural networks. In this paper， we analyze the aggressiveness of SAR adversarial samples， and the difference between SAR adversarial examples and original examples in the frequency domain. With the analysis results， a two-step SAR adversarial samples detection technique is proposed to improve the security of SAR classification models. The first step of adversarial sample detection is performed on the input image based on the frequency domain analysis to separate the adversarial samples. Then， the remaining images are fed into an adversarial trained model and an untrained model to complete the second step of adversarial sample detection. By using this two-step detection method， the adversarial samples can be effectively detected with a detection success rate of no less than 95.73%， effectively improving the security of the SAR classification model.

Abstract:An multi-scale synthetic aperture radar （SAR） image detection algorithm for ships based on improved YOLOv5 is proposed to address the large pixel scale difference of ship targets in complex scenes and missed detection caused by dense array of ships. For the neck network of YOLOv5， a bi-directional feature pyramid network （BiFPN） is adopted to enhance the multi-scale feature fusion ability of the network， and an enhanced channel-MLP （EC-MLP） module is constructed based on depthwise separable convolution （DSC） and channel MLP in its bottom-up feature fusion branch to enrich semantic information and provide more sufficient ship target context features. The global attention mechanism （GAM） is introduced to enable the network to extract input features selectively and reduce information reduction. In addition， the SIoU loss function is used to further improve the training convergence speed and detection accuracy of the network. Comparative experiments with eight other methods （Faster R-CNN， Libra R-CNN， FCOS， YOLOv5s， PP-YOLOv2， YOLOX-s， PP-YOLOE-s and YOLOv7-tiny） are conducted on SSDD and HRSID datasets. The experimental results show that the AP50 of the improved algorithm reaches 96.7% on SSDD and 95.6% on HRSID， which is superior to the comparison methods.

Abstract:This paper presents a roll angle calibration method for phased-array radar. Based on the built-in elevation calibration methods of phased-array radar， the proposed method utilizes the approximately linear relationship between the roll angle and the target’s elevation measurement error to calculate the roll angle. Experiments confirm that the roll angle calculated by this method has nearly the same accuracy as that of the method using laser measurement instruments. With the proposed method， the angle measurement accuracy of phased-array radar can be greatly improved.

Abstract:This paper combines simultaneous transmitting and reflecting reconfigurable intelligent surfaces （STAR-RIS） with integrated sensing and communication （ISAC） systems to achieve full space communication and awareness. At the same time， a low-cost sensor is applied to STAR-RIS to achieve target sensing on STAR-RIS， solving the serious path loss problem of radar sensing. Based on this， this article researches ISAC system of STAR-RIS assisted multi user multi input single output （MU-MISO） located on both sides of STAR-RIS and a target located on the transmission side of STAR-RIS， aiming to jointly design the active beamforming at the ISAC base station and the passive beamforming matrix of STAR-RIS， in order to maximize communication sum-rate， At the same time， it meets the minimum signal to noise ratio requirements for target perception performance. To solve the non-convex problem in the optimization process， this paper proposes a block coordinate ascending algorithm based on fractional programming， which divides the optimization variables into several block variables for alternate optimization. In the subsequent active and passive beamforming problems of iterative optimization， efficient algorithms such as continuous convex approximation and semi definite relaxation are applied. Simulation results validate the advantages of deploying STAR-RIS in ISAC systems compared to traditional reconfigurable intelligent surfaces. At the same time， the proposed fractional programming-based algorithm is compared with the weighted minimum mean square error algorithm. The simulation results verify the advantages and effectiveness of the proposed algorithm in improving the sum-rate of communication.

Abstract:Considering the problem of unsupervised domain adaption semantic segmentation， it is very important to establish a long-distance context relationship between the source domain and the target domain and how to solve the problem of unbalance distribution of different classes of pixels. we propose a dual cross-domain graph convolution network to exploit the long-distance context between source and target domain and fuse the feature of two domains. Specifically， we construct the position similarity matrix and channel similarity matrix of the cross domain and propose the cross-domain position graph convolution and cross-domain channel graph convolution. In order to solve the problem of unbalanced distribution of classes in the datasets and capture more domain invariant feature， we propose a group contrastive learning strategy to narrow the distance between the same class of two domains and widen the distance between the different classes of two domains by constructing positive and negative samples in the group. A large number of experiments show that our method achieves good performance on Urban Scene datasets GTA5 to Cityscapes and SYNTHA to Cityscapes.

Abstract:Due to the geometric distortion of the document paper， the interference from the shooting scene， and perspective distortion brought on by the unfavorable shooting angle， the optical character recognition （OCR） quality of document photos taken by mobile devices has been severely hampered. Two networks based on auto-encoder are created to perform adaptive image correction and increase the accurate rate of text recognition in order to handle pre-processing distorted document images with folding and distortion. First， we propose two different types of residual blocks： dilated residual blocks and asymmetric convolutional residual blocks， and then combine the residual blocks with the auto-encoder to create an asymmetric dilated auto-encoder. In the meantime， we create a spatial pyramid auto-encoder by using spatial pyramid pooling instead of fully connected layers and implementing feature extraction with asymmetric convolutional residual blocks. Experimental results show that， compared with distorted images， the corrected images by the asymmetric dilated auto-encoder respectively improve by 26.3%， 20.4% and 12.3% in OCR precision， OCR recall， and text similarity. Besides the corrected images by the spatial pyramid auto-encoder respectively improve by 27.7%， 22.0% and 15.5% in OCR precision， OCR recall， and text similarity. Compared with other image rectification networks such as RectiNet， the corrected images by these two auto-encoders perform much better on optical character recognition. The corrected document images of both asymmetric dilated auto-encoder and spatial pyramid auto-encoder are effectively improved in terms of OCR precision， OCR recall， and text similarity. Not only that， they have relatively obvious advantages over existing networks in terms of robustness and generalizability.

Abstract:Network intrusion detection system is an important defense tool for network information security protection， and the complicated and lengthy network intrusion behavior features seriously affect the effectiveness of network intrusion detection. In order to solve the problem of rapid information growth and incomplete data in network intrusion detection， an incremental feature selection algorithm based on neighborhood valued tolerance condition entropy is proposed. Firstly， on the basis of neighborhood valued tolerance granular computing， combined with the remarkable characteristics of conditional entropy in characterizing the uncertainty of features and the correlation or dependency between features， the incremental updating mechanism of neighborhood valued tolerance conditional entropy is studied. Then， based on the update mechanism， an incremental feature selection algorithm for dynamic database is proposed. Finally， the experimental analysis shows that the proposed algorithm can effectively improve the computational efficiency of feature selection in incomplete information systems. The new algorithm has the advantages of low computational complexity and low false alarm rate in the application of network intrusion detection examples， which shows that it can provide effective and feasible concrete methods for network information security protection.

Abstract:Traditional models suffer from feature sparsity， feature loss and incomplete comment feature extraction problems due to the imbalance of comment length. This paper proposes an emotional analysis approach based on dynamic word-sentence features and self-attention （DWSF-SA）， to alleviate the incomplete extraction problem caused by the imbalance of text size under batch training. DWSF-SA first follows pre-training on dynamic feature embedding， then employs sentence vectors to complete the less parts and represents the truncated parts by fixed length. Moreover， DWSF-SA also introduces a self-attention mechanism to dynamically integrate the word-sentence fusion features， and makes optimization on the weight parameters to accelerate the computation and training. The ablation and comparison experiments on publicly available datasets demonstrate that the proposed DWSF-SA outperforms traditional approaches in accuracy metrics.

Abstract:Weakly supervised video anomaly detection has become a hot spot in video anomaly detection research due to its strong anti-interference and low data labeling requirements. In the existing methods， most of the weakly supervised video anomaly detection methods assume that the clips in each video distribute independently， and determine whether it is abnormal for each video clip independently， ignoring the temporal and spatial information between video clips. To alleviate these problems， this paper proposes a weakly supervised anomaly detection method based on spatio-temporal dependence and feature fusion. Retaining the original characteristics of video clips， this method uses the distance of index and the similarity of features between video clips to fit the time dependence and the spatial dependencies of video， which builds the relationship characteristics of video clips. By fusing the original features and relationship features， the dynamic characteristics and temporal relationship of videos can be better expressed. Extensive experiments on two benchmark datasets， UCF-Crime and ShanghaiTech， demonstrate that the proposed method outperforms other methods with the AUC values reaching 80.1% and 94.6%， respectively.

Abstract:In large-scale resource environments， traditional resource indexing mechanisms lead to a rapid increase in the number of Peer nodes and a decrease in load balancing performance， affecting query efficiency and system stability. This paper introduces a centroid model-based local resource clustering method， which clusters similar resources at a single node and selects a representative key value， effectively reducing the scale of Peer nodes in the peer-to-peer（P2P） network. Additionally， the local clustering mechanism focuses on processing closely related key values， thus preventing excessive expansion of resource coverage. Experimental results demonstrate that the Skip Graph algorithm based on the centroid model not only reduces query complexity and improves load balancing performance， but also exhibits excellent scalability in terms of network size， data volume， and query complexity， better adapting to the needs of large-scale resource discovery.

Abstract:Automatic and accurate polyp localization and segmentation methods can detect polyps in a timely manner in the early stage of colorectal cancer lesions， greatly reducing the risk of cancer transformation. The encoder-decoder architecture， as the most mainstream network structure in polyp segmentation in recent years， has been greatly improved， such as improving the model’s ability to capture global contextual and local features， and using deep features to guide shallow decoding. However， polyps vary in shape and size， and due to their convolutional nature， they are prone to getting too caught up in local information mining and losing remote information dependencies during encoding. Some polyp images also have low contrast and complex spatial characteristics， which makes it easy to confuse the polyp with the background. Based on this， this paper proposes a polyp segmentation network based on multiple attention and schatten-p norm（MASNet）. Among them， the axial multiple attention module utilizes axial attention to supplement remote contextual relationships in the image， while also paying attention to boundary and background information to achieve feature complementarity. It enhances the capture of local detail features while paying attention to global features. By utilizing the correlation between matrix singular values and matrix implicit information， the schatten-p norm is introduced as a constraint to analyze the data from a matrix perspective and assist the model in distinguishing foreground and background. By setting up a large number of experiments， the effectiveness of the proposed method is proven， and MASNet achieves the best segmentation results by comparing different advanced methods on the Kvasir-SEG dataset.

Abstract:The development and application of pulse diagnostic instruments provide an objective basis for the intelligent diagnosis of traditional Chinese medicine. However， the existing pulse diagnostic instruments do not consider the influence of the collection region （Cun， Guan， Chi） and pressure （Fu， Zhong， Chen） on the diagnostic results， and there is still room for the improvement of the diagnostic accuracy. In order to recognize pulse condition more accurately， this paper presents an intelligent pulse diagnosis system based on flexible sensors and the corresponding pulse signal processing method. By installing three array flexible sensors at the collection region of Cun， Guan and Chi and setting different pressure thresholds of Fu， Zhong and Chen， multiple pulse signals are obtained. Signal features are then extracted， and multi-channel features are integrated based on multi-set canonical correlations analysis （MCCA） to get more pulse information. Experimental results show that the proposed method can further improve the accuracy of pulse condition classification in four typical pulse types. The multi-point pulse condition induction designed in this paper based on two aspects of region and pressure can simulate and restore the real Chinese medicine diagnosis process and help to extract real pulse signals， providing a theoretical basis and reference value for the subsequent research and development of intelligent pulse diagnosis instruments based on flexible sensors.