Cluster DetectionMethod of Endogenous Security Abnormal Attack Behavior in Air Traffic Control Network

In order to enhance the accuracy of Air Traffic Control(ATC)cybersecurity attack detection,in this paper,a new clustering detection method is designed for air traffic control network security attacks.The feature set for ATC cybersecurity attacks is constructed by setting the feature states,adding recursive features,and determining the feature criticality.The expected information gain and entropy of the feature data are computed to determine the information gain of the feature data and reduce the interference of similar feature data.An autoencoder is introduced into the AI(artificial intelligence)algorithm to encode and decode the characteristics of ATC network security attack behavior to reduce the dimensionality of the ATC network security attack behavior data.Based on the above processing,an unsupervised learning algorithm for clustering detection of ATC network security attacks is designed.First,determine the distance between the clustering clusters of ATC network security attack behavior characteristics,calculate the clustering threshold,and construct the initial clustering center.Then,the new average value of all feature objects in each cluster is recalculated as the new cluster center.Second,it traverses all objects in a cluster of ATC network security attack behavior feature data.Finally,the cluster detection of ATC network security attack behavior is completed by the computation of objective functions.The experiment took three groups of experimental attack behavior data sets as the test object,and took the detection rate,false detection rate and recall rate as the test indicators,and selected three similar methods for comparative test.The experimental results show that the detection rate of this method is about 98%,the false positive rate is below 1%,and the recall rate is above 97%.Research shows that this method can improve the detection performance of security attacks in air traffic control network.

Performance Metrics and Delay Thresholds for Multi-Link Cluster Monitoring in Seismic Profession

Seismic cluster nodes can be monitored by monitoring system,but thresholds for link failure alarm in monitoring systems are not determined presently,especially in different types of cluster links.Communication link types are discussed in seismic profession.By analyzing the characteristics of various links,the main performance metric,network latency,was proposed,which influenced states of communication links and gave the monitoring results deviation formula for judging the cluster monitoring system at different delay thresholds settings based on multiple-link delay error ratio analysis algorithm we offered.From the final experimental data of the monitoring system,fault alarm thresholds settings were posed under five different communication links,which had the instruction significance to the cluster monitoring in seismic profession.

Early warning system for shallow landslides using rainfall threshold and slope stability analysis

A combined cluster and regression analysis were performed for the first time to identify rainfall threshold that triggers landslide events in Amboori, Kerala, India. Amboori is a tropical area that is highly vulnerable to landslides. The 2, 3, and 5-day antecedent rainfall data versus daily rainfall was clustered to identify a cluster of critical events that could potentially trigger landslides. Further, the cluster of critical events was utilized for regression analysis to develop the threshold equations. The 5-day antecedent(xvariable) vs. daily rainfall(y-variable) provided the best fit to the data with a threshold equation of y = 80.7-0.1981 x. The intercept of the equation indicates that if the 5-day antecedent rainfall is zero, the minimum daily rainfall needed to trigger the landslide in the Amboori region would be 80.7 mm. The negative coefficient of the antecedent rainfall indicates that when the cumulative antecedent rainfall increases, the amount of daily rainfall required to trigger monsoon landslide decreases. The coefficient value indicates that the contribution of the 5-day antecedent rainfall is～20% to the landslide trigger threshold. The slope stability analysis carried out for the area, using Probabilistic Infinite Slope Analysis Model(PISA-m), was utilized to identify the areas vulnerable to landslide in the region. The locations in the area where past landslides have occurred demonstrate lower Factors of Safety(FS) in the slope stability analysis. Thus, rainfall threshold analysis together with the FS values from slope stability can be suitable for developing a simple, cost-effective, and comprehensive early-warning system for shallow landslides in Amboori and similar regions.

SLID: A Secure Lowest-ID Clustering Algorithm

In order to solve security problem of clustering algorithm, we proposed amethod to enhance the security of the well-known lowest-ID clustering algorithm. This method isbased on the idea of the secret sharing and the (k, n) threshold cryptography, Each node, whetherclusterhead or ordinary member, holds a share of the global certificate, and any k nodes cancommunicate securely. There is no need for any clusterhead to execute extra functions more thanrouting. Our scheme needs some prior configuration before deployment, and can be used in criticalenvironment with small scale. The security-enhancement for Lowest-ID algorithm can also be appliedinto other clustering approaches with minor modification. The feasibility of this method wasverified bythe simulation results.

A Low Energy Uneven Cluster Protocol Design for Wireless Sensor Network

A low energy uneven cluster protocol design method is proposed. Aiming at the random choosing for cluster head of traditional Leach protocol, and the defect of the single hop from all the cluster heads to the sink node, an improved method for Leave protocol is advanced. Firstly, the election model of cluster head is improved, and the node residual energy is considered in the process of threshold and the cluster head election to improve the whole network life circle. In the multi-hop route, choosing the maximum energy and the nearest node as the next hop and a route transferring data among many clusters is formed. The experiment shows the method having great improvement compared with Leach protocol and prolonging the network life cycle.

Effective Cluster Head Substitution Leach Protocol for Wireless Communication

In recent years, the demand for Wireless Sensor Network (WSN) in smart farming has had a tremendous increase in demand for its efficiency. Wireless sensor networks have very many nodes, and it is of no use when the battery dies. This is why there are several routing protocols being take into consideration to cub this problem. In this paper, in order to increase the heterogeneity and energy levels of the network, the M-LEACH protocol is proposed. The key aim of the Leach protocol is to prolong the existence of wireless sensor network by lowering the energy consumption needed for Cluster Head creation and maintenance, the proposed algorithm instructs a node to use high power amplification as it acts as the Cluster heads, and low power amplification when it becomes a Cluster Member, in the next stage. Finally, for better effectiveness, M-LEACH employs hard and soft threshold systems. Since it eliminates collisions and reduces the packet drop ratio for other signals, the M-LEACH protocol proposed works better than the Leach protocol.

诱发台风暴雨型滑坡的降雨阈值研究——以泰顺县为例

在我国东南沿海地区,台风暴雨诱发了大量山体滑坡。建立诱发台风暴雨型滑坡的降雨阈值模型,可为该地区的滑坡预警提供参考依据。基于浙江省泰顺县2007—2022年以来发生的滑坡信息和日降雨量数据,进行了诱发滑坡的台风暴雨事件编目,初步揭示了台风暴雨型滑坡的发育规律和触发降雨特征,通过降雨强度(I)、降雨历时(D)的幂律关系构建了滑坡临界降雨阈值模型,并开展了不同台风降雨雨型的阈值模型对比。结果表明:台风暴雨型滑坡大量发生在与台风登陆方向接近的迎风坡,坡表植被多为乔木和竹林;诱发滑坡的台风降雨一般持续2~3 d,且降雨过程集中,降雨雨型以单峰型为主,总降雨量一般在200 mm以上;台风暴雨诱发滑坡的降雨阈值显著高于一般性的降雨,这种差异与台风降雨模式及大尺度气候环境有关;雨型也会显著影响降雨阈值,随着降雨峰值的后移,阈值逐渐降低,表明台风暴雨型滑坡对长时间降雨后出现的强降雨事件更加敏感。通过降雨阈值模型对9次台风降雨事件是否诱发滑坡进行了预测,预测结果与实际情况比较吻合,证明该模型及研究思路对台汛期东南沿海的滑坡监测预警有较强的参考价值。

基于网格近邻优化的密度峰值聚类算法

密度峰值聚类(DPC)将数据样本点的局部密度和相对距离进行结合,能对任意形状数据集进行聚类处理,但密度峰值聚类算法存在主观选择截断距离、简单分配策略和较高时间复杂度等问题。为此,提出了一种基于网格近邻优化的密度峰值聚类算法(KG-DPC算法)。首先对数据空间进行网格化,减少了样本数据点之间距离的计算量;在计算局部密度时不仅考虑了网格自身的密度值,而且考虑了周围k个近邻的网格密度值,降低了主观选择截断距离对聚类结果的影响,提高了聚类准确率,设定网格密度阈值,保证了聚类结果的稳定性。通过实验结果表明,KG-DPC算法比DBSCAN、DPC和SDPC算法在聚类准确率上有很大提升,在聚类平均消耗时间上DPC、SNN-DPC和DPC-NN算法分别降低38%、44%和44%。在保证基本聚类准确率的基础上,KG-DPC算法在聚类效率上有特定优势。

Thresholds of Instability: Precipitation, Landslides, and Early Warning Systems in Brazil

Rainfall accumulation thresholds are crucial for issuing landslide warnings by identifying when soil saturation from rain could potentially trigger a landslide. Two essential types of thresholds are considered: environmental and operational. The environmental threshold indicates the minimum rainfall level required to potentially initiate a landslide. Conversely, the operational threshold is set lower to enable agencies to issue alerts before reaching environmental thresholds. Establishing these thresholds improves the accuracy of landslide predictions in terms of location and timing. This study introduces an innovative approach for determining these thresholds. Our approach employs cluster analysis and historical landslide data from the Metropolitan Region of Recife, Pernambuco State, Brazil. We applied our defined values to a significant landslide event in 2022, validating their robustness as the foundation for the operational threshold used by Cemaden, Brazil’s National Center for Monitoring and Early Warning of Natural Disasters.

A Support Construction for CT Image Based on K-Means Clustering

Computer Tomography in medical imaging provides human internal body pictures in the digital form. The more quality images it provides, the better information we get. Normally, medical imaging can be constructed by projection data from several perspectives. In this paper, our research challenges and describes a numerical method for refining the image of a Region of Interest (ROI) by constructing support within a standard CT image. It is obvious that the quality of tomographic slice is affected by artifacts. CT using filter and K-means clustering provides a way to reconstruct an ROI with minimal artifacts and improve the degree of the spatial resolution. Experimental results are presented for improving the reconstructed images, showing that the approach enhances the overall resolution and contrast of ROI images. Our method provides a number of advantages: robustness with noise in projection data and support construction without the need to acquire any additional setup.

基于遗传算法的简易优化及阈值确定研究

为了保持原始功能磁共振成像(function Magnetic Resonance Imaging, fMRI)数据的空间结构实现噪声体素移除,提高聚类的效果,提出了一种基于遗传算法的简易优化算法.以广义线性模型(Generalize Linear Model, GLM)方法获取的数据作为大脑视觉刺激的真实激活模板,基于遗传算法对5位被测者的fMRI数据图像进行分析.取不同阈值(0.1~0.9)时交叉率和变异率为0且具有唯一的最优值为模糊C均值聚类算法(Fuzzy C-Means, FCM)结果,依据真实模板验证聚类结果的准确度.结果表明,相比原始的FCM方法,改变阈值的大小可以使FCM聚类结果的准确度得到有效提高.通过简易优化遗传算法,可以确定最佳阈值为0.6.

基于FCM及快速迭代收缩阈值算法的平面ECT图像重建

为提高平面阵列电容成像系统的成像精度,提出一种基于模糊C均值聚类(FCM)进行数据优化的快速迭代收缩阈值算法(FISTA)。根据平面阵列电容数据的特点,首先利用FCM算法对测量电容值进行分类,保留有效电容值,实现电容向量降维;然后利用离散小波基(DWT)对灰度值进行稀疏表示,并建立L1正则化模型,采用FISTA进行求解,以实现图像重建;最后将FCM处理后的电容值分别用于Landweber算法、Tikhonov算法进行重建对比。仿真与实验结果表明,该算法重建图像的平均相对误差约为0.0527,平均相关系数约为0.9422,均优于其它算法,且重建图像伪影较少,更接近真实情况;因此,所提算法具有更好的重建效果。。

红外图像分割与目标提取方法研究

近年来红外热像仪在军事和安防领域广泛使用,然而受环境因素的影响和红外热像仪器件本身制造工艺的限制,其成像的效果并不理想。本文主要研究了阈值分割、算子边缘检测、均值漂移聚类分割算法等几种图像分割方法,并采用MATLAB软件进行了仿真,通过对仿真结果的分析与对比,确定了对不同类型红外图像进行图像分割时选取分割算法的原则,给出了每类红外图像所适用的分割算法,并对视频图像中的目标进行了提取与跟踪。

基于时变性聚团识别方法的高密度循环流化床内聚团分布特性

为研究高密度提升管内的聚团分布特性,在18 m高循环流化床提升管系统中测定了不同固体循环速率下(G_(s)=100~1800 kg·m^(-2)·s^(-1))的瞬时颗粒浓度。采用基于时变性的聚团识别方法对聚团进行识别,进而得到了聚团浓度、聚团持续时间和聚团速度在提升管中的分布。研究结果表明,高密度循环流化床中的颗粒浓度和聚团特性参数和与低密度循环流化床相比存在较大差异。高密度条件下,提升管内聚团浓度最高可达0.41。聚团持续时间最高可达5.5 ms。提升管截面内的聚团净向上运动,r/R=0.949时,聚团的净运动速度高达1.83 m·s^(-1)。此外,在极高的颗粒循环速率下(G_(s)=1800 kg·m^(-2)·s^(-1)),聚团特性参数的轴径向分布变得更均匀。

基于三支决策的灰色可能度聚类方法及应用

针对经典的灰色可能度聚类评估模型难以判定决策对象的灰类归属和过度聚类等问题,利用三支决策的思想和方法,通过引入三支灰类的概念描述决策对象和灰类之间的不确定聚类关系;将其代替灰色定权聚类中的灰类和严格的聚类关系,构建基于三支决策的灰色可能度聚类方法,并采用决策粗糙集中的贝叶斯推理确定聚类阈值;最后,以案例验证所提方法的有效性和合理性。结果表明:本文所构建的模型是经典灰色可能度聚类评估模型的拓展和泛化,可以有效避免过度聚类,降低决策风险,提高聚类可靠性。

聚类不确定性的三支决策方法综述

由于数据有复杂性、多样性和不确定性的特点,聚类通常存在不确定性,这种不确定性是指聚类结果的不一致性,从而影响分析的准确性,而通过三支决策方法构建不确定域,对不确定域进行延迟决策,能够带来一系列的优势。近些年来,这个研究热点涌现许多重要的工作。对聚类不确定的三支决策方法模型和研究方向进行介绍,按照三支聚类构建过程中遇到的问题进行总结。首先,给出三支聚类的表现形式。其次,分别阐述如何构建三支聚类的核心域和边界域,如何确定阈值,如何设计有效性指标,并介绍其中具有代表性的工作。最后,对聚类不确定的三支决策方法进行总结,并展望未来的研究方向。

基于LEACH协议的能量优化方法

针对低功耗自适应聚类协议产生的整体能耗不均及网络生命周期短的问题,提出一种改进的能量优化方法。结合能量模型、节点发送及融合数据的综合能耗计算簇头比例;利用节点剩余能量与节点基站间的距离,优化成簇阶段中簇头选取阈值公式。在普通节点入簇阶段,引入簇规模密度,避免簇规模差距过大,采用就近原则与簇密度规模更新簇头。在数据传输阶段,引入“热区”边界对簇头分类,不同类簇头采用不同的数据传输方式。仿真结果表明,所提方法能均衡网络能耗且延长网络生命周期。

无线传感器网络中一种基于聚合层次聚类的分簇路由算法

针对无线传感器网络中节点连接以及能量受限不足的问题,为了延长网络寿命,提出了一种基于AHC的分簇路由算法(HACCRA)。该算法首先运用AHC对网络节点分簇,接着为簇首选择、簇形成和路径构建分别定义了恰当的决策目标函数,运用能量阈值、提出距离阈值、并且路由过程优先考虑簇首节点之间的一对一连接,有效解决了路由算法中分簇和路由不衔接的问题。仿真结果表明,与JCR、ICR以及DCK-LEACH相比,HACCRA能够更好地实现网络节点的能耗均衡,保证网络数据传输的连接性,从而延长网络寿命。

格式塔原则下的建筑群空间分布智能理解

在格式塔原则下对建筑群进行聚类是计算机理解建筑群空间分布特征并实现地图自动综合的基础.在格式塔原则约束下,首先,利用最小生成树的阈值聚类及参数聚类,实现了对集中和分散均匀排列建筑群空间分布特征的挖掘,并利用C_(v)^(-)值定量评价了挖掘质量;其次,提出了阈值聚类和参数聚类的异同,以及阈值聚类具有局限性的原因;最后,根据建筑物语义特征获得最终聚类结果.试验表明:1)传统的MST阈值聚类不能维持分散均匀排列建筑群的空间分布特征;2)MST参数聚类的适应性要高于阈值聚类,更加符合人类的视觉认知.该方法更有助于提高计算机对建筑群空间分布的智能理解能力.

无线传感网LEACH协议的改进方法

为解决传统LEACH(Low Energy Adaptive Clustering Hierarchy)协议网络节点能量消耗高、存活数量少和生存寿命短等问题,提出了一种LEACH-AD改进方案。该算法引入最优簇头比率P值、加入距离因子、剩余能量因子和密度因子等因素更新的阈值公式进行分簇以及簇间的传输。实验结果表明,改进后的LEACH-AD协议在首个死亡节点、10%死亡节点以及全部死亡节点分别比原LEACH协议延长138轮、195轮、628轮。在能量消耗方面比原LEACH协议多持续了631轮,改进后的路由协议减少了网络节点的能量消耗量,从而有效延长了无线网络与传感节点的工作时间,这对无线监测系统的研究与开发意义重大。