Multi-Featured and Fuzzy Based Dual Analysis Approach to Optimize the Subspace Clustering for Images

Wireless Personal Communications - In unsupervised classification, the subspace clustering is gaining the scope for the categorization of the more comprehensive and random image pool. In this...     

An Alternative Queuing Technique to Optimize the Performance of TDMA Based Mobile Communication Systems

This paper analyses an alternative traffic queuing model for the quality of service (QoS) optimization in mobile communications systems. According to the proposed model, the queue of handover attempts is separated to each of the used transceivers of a base transceiver station that covers a particular area. Fixed channel assignment and TDMA are considered, while the proposed technique is compared to the classic queuing model. The comparison results show that the QoS of the proposed model is fully optimized, especially if large queue sizes are chosen.     

Enhancing Clustering Stability in VANET: A Spectral Clustering Based Approach

Vehicles can establish a collaborative environment cognition through sharing the original or processed sensor data from the vehicular sensors and status map. Clustering in the vehicular ad-hoc network(VANET) is crucial for enhancing the stability of the collaborative environment. In this paper, the problem for clustering is innovatively transformed into a cutting graph problem. A novel clustering algorithm based on the Spectral Clustering algorithm and the improved force-directed algorithm is designed. It takes the average lifetime of all clusters as an optimization goal so that the stability of the entire system can be enhanced. A series of close-to-practical scenarios are generated by the Simulation of Urban Mobility(SUMO). The numerical results indicate that our approach has superior performance in maintaining whole cluster stability.     

VANET中基于协议序列及TDMA的混合信道接入机制

为了提高高速公路场景下用户平均有效数据吞吐量,提出车载自组织网络分布式信道接入机制.通过道路区段划分实现协议序列集复用;通过引入少量反馈并利用UI(User Irrepressibility)序列特性实现节点间接入顺序感知;同区段节点间利用分布式TDMA(Time Division Multiple Access)模式最大化吞吐量.仿真结果表明,其吞吐量性能优于经典的基于GNSS(Global Navigation Satellite Systems)的机制,并且物理层数据传输速率越大,节点通信半径越大,车辆密度越稀疏,其优势越明显.     

A Hybrid Task Scheduling Algorithm Based on Task Clustering

Mobile Networks and Applications - Based on the problem of task communication overhead being higher than the task execution time has a direct negative impact on the makespan of task scheduling in...     

A TDMA Approach for OFDM-Based Multiuser RadCom Systems

This paper is focused on the multiuser implementation of fusion of radar and communication(RadCom) in internet-of-vehicles(IoV) scenarios.Traditional time-division multiple access(TDMA)technology degrades the velocity detection performance of orthogonal frequency-division multiplexing(OFDM)-based RadCom systems. We propose a new TDMA approach for OFDM-based RadCom systems, where multiuser communication and radar detection are completed synchronously. We consider a continuous-wave TDMA OFDM struc...     

基于PRI熵的雷达信号聚类方法研究

针对脉冲分选聚类中存在的问题,引入熵的概念,定义了 PRI(脉冲重复间隔)熵,给出了雷达脉冲序列之间合并的 PRI 熵判据.并在此基础上提出了新的雷达信号聚类方法.采用先分选后合并的思路,通过计算合并前后 PRI 熵是否减小来判断是否对序列合并.通过仿真验证该方法解决增批问题的有效性.     

A Genetic Algorithm Approach to Optimize Parameters in Infrared Guidance System

In the infrared guidance system, the gray level threshold is key for target recognition. After thresholding, a target in the binary image is distinguished from the complex background by three recognition features. Using a genetic algorithm, this paper seeks to find the optimal parameters varied with different sub-images to compute the adaptive segmentation threshold. The experimental results reveal that the GA paradigm is an efficient and effective method of search.     

一种新的基于混合蚁群算法的聚类方法

建立了聚类分析问题模型，分析了K-均值算法、模拟退火算法和基本蚁群算法的优缺点。对蚁群算法作了改进．思路是K-均值方法混合，利用K-均值方法的结果作为初值。经过比较测试，两种混合蚁群算法的效果都比较好．特别混合方法二的效果最好。     

视觉采样聚类方法VSC

基于视觉采样原理,该文提出了一般化的视觉采样聚类方法VSC。该方法将视觉原理与著名的Weber定律结合起来,其特点是：对聚类初始条件不敏感;Weber定律提供了新的聚类有效性标准,并且该方法所得到的合理的聚类数可以依据Weber定律而得到。大量的实验结果表明了算法VSC的有效性。文中讨论了算法VSC与由Yang Miin-Shen等人(2004)新近提出的基于相似度量的聚类算法SCA之间的内在联系,得出了这两个算法具有一定的同解性质,从而揭示了该文所提方法VSC能够有效地克服算法SCA中参数 不易确定的困难。     

基于蚁群和模拟退火算法的聚类新方法

模拟退火算法具有良好的全局搜索能力,而蚁群算法具有良好的分布式并行性和正反馈能力.针对样本维数大、数目多时聚类效果不满意的问题,提出了混合的蚁群模拟退火算法,思路是利用K-均值算法的结果作为初值,再使用蚁群算法和模拟退火算法对初值进行调整聚类,结果表明这种算法比较有效.     

基于自适应门限MDL方法的TDMA突发特征码检测

阐述了基于最小描述长度(MDL)方法来进行特征码检测,根据检测量分布的推导结果,提出了检测门限对检测性能、特征码位置不确定性和信噪比变化自适应的MDL特征码检测方法。当基于特征码进行通信信号分选时,新方法具有更好的应用灵活性,且检测性能对频偏具备鲁棒性。     

一种新的快速混合聚类算法

在汲取了传统划分、层次聚类方法优点的基础上,结合图搜索技术,提出了一种新的快速混合聚类算法.该算法主要分为三个步骤:首先将整个数据集"压碎",生成固定数量的原子级聚类;然后处理孤立点;最后采用图搜索技术生成聚类.该算法只需一个参数,能识别任意形状、大小的聚类,时间复杂度在最坏情况下为nn~(1/2).实验表明该算法是有效的.     

求解聚类问题的混合PSO算法设计

提出了一种基于混沌理论的求解聚类问题的混合PSO算法,该算法结合相关算法的优势采用混沌扰动生成初始化聚类种子,以保证解的分布性能;采用PSO算法进行初步聚类,以改善算法的全局搜索性能;然后采用K-均值算法进行局部精练,以加快算法的收敛速度。将粒子分为静态和动态两种,精心设计了动态粒子的属性,用碟型数据和IRIS数据仿真实验表明,该算法是有效的,能很好地解决聚类问题。     

基于混合聚类算法的客户细分策略研究

针对层次聚类法和 K-means 聚类法的缺陷和不足,提出将二者相结合的改进算法,既解决了层次聚类法伸缩性差的问题,又解决了 K-means聚类法对初始聚类中心敏感的问题。通过对改进算法的计算复杂度分析并利用 UCI 数据库的测试数据对改进算法进行测试。结果表明,混合聚类算法使样本聚类的准确率提高到94%,并有更高的执行效率和更好地实用性。此外,将此算法应用到汽车销售公司的客户细分管理中,得出了差别化明显的客户细分类别,表明此改进算法具有更强的客户细分能力以及客户行为特征的解释能力。     

A Practical Optimization Approach for QPSK/TDMA Satellite Channel Filtering

This paper reports the results of computer and laboratory hardware simulation studies concerning optimum pulseshaping filter characteristics and configurations to be employed with conventional QPSK signaling in a nonlinear channel. An understanding of the factors and issues contributing to transmission impairments was obtained through an initial hardware simulation which was confirmed and extended with computer simulation. Based on this understanding, the critical characteristics of the channel pulseshaping transmit and receive filters were determined, promising filter candidates were implemented, and a complete system laboratory hardware simulation study was configured to evaluate the candidate filters using an INTELSAT IV satellite. This study emphasizes the performance tradeoff between the nonlinear channel-induced bit-error-rate impairment and the adjacent channel interference levels caused by HPA spectrum regeneration of filtered QPSK as a function of filter configuration. The results show that a more gentle rolloff Nyquist or wider bandwidth non-Nyquist transmit filter produces similar significantly improved performance compared to a conventional sharp rolloff Nyquist transmit filter in nonlinear satellite channels. For the wider non-Nyquist filter case, employing a Nyquist filter at the demodulation also benefits performance.     

基于聚类的少数类样本采样方法

针对处理不平衡数据集的分类问题,SMOTE通过在相邻样例间线性插值实现少数类样例过采样。但SMOTE插值的结果是样例密集的地方依然相对密集,样例稀疏的地方依然相对稀疏,影响分类性能。针对该问题本文提出一种基于聚类的过采样方法-C-SMOTE。该方法首先将少数类样例聚成多个簇,再以簇为单位结合SMOTE方法产生新样例。实验结果表明,C-SMOTE既保证了数据集整体分类准确率,又能提高少数类分类精度。     

A Novel Approach to Optimize Power Consumption in Orchard WSN: Efficient Opportunistic Routing

Wireless Personal Communications - In orchard wireless sensor networks (WSN) application, the network link quality is unstable due to the shielding effect caused by the growth of branches and...     

基于TDMA的视讯综合应用平台

视讯综合应用平台是以视频监控、视讯会议、远程教育、事故抢险为主要功能的综合业务系统,采用基于TDMA的网络通信方案是本系统的主要特点。文章在深入研究TDMA技术的基础上,讨论了此方案的优点与必要性,并介绍了实现网络信道的动态时隙分配、设备控制、远程视频传输的方案,给出了相应的算法举例。该产品在铁路、高速公路、长途客运均有广泛的应用前景。     

A Clustering TDMA and Code Block Binding Algorithms to Enhance the Node Energy and Security of the Network

Wireless Personal Communications - Wireless body area Networks is used to measure the ambient conditions related to the environment surrounding the sensors and transforms them into an electrical...