为解决密集Wi-Fi网络中重叠基本服务集(OBSS,overlapping basic service set)用户受到严重的同频干扰,导致接收信号信干噪比(SINR,signal to interference plus noise ratio)值低、数据传输速率低或者无法传输的问题,提出了一种基于改...为解决密集Wi-Fi网络中重叠基本服务集(OBSS,overlapping basic service set)用户受到严重的同频干扰,导致接收信号信干噪比(SINR,signal to interference plus noise ratio)值低、数据传输速率低或者无法传输的问题,提出了一种基于改进遗传算法(IGA,improved genetic algorithm)选择接入点(AP,access point)协作集的多AP联合传输(JT,joint transmission)方案。首先,利用SINR阈值法将系统内所有站点(STA,station)分为中心用户和边缘用户,边缘用户采用JT方案。然后,以最大化边缘用户吞吐量为目标,利用IGA为每个边缘用户确定合适的协作AP形成其专属的AP协作集,最大化JT技术的潜在增益。仿真结果表明,所提方案在密集Wi-Fi网络场景下能够有效减小基本服务集(BSS,basic service set)间的同频干扰,提升BSS边缘用户的性能和吞吐量。展开更多
Affinity propagation(AP)is a classic clustering algorithm.To improve the classical AP algorithms,we propose a clustering algorithm namely,adaptive spectral affinity propagation(AdaSAP).In particular,we discuss why AP ...Affinity propagation(AP)is a classic clustering algorithm.To improve the classical AP algorithms,we propose a clustering algorithm namely,adaptive spectral affinity propagation(AdaSAP).In particular,we discuss why AP is not suitable for non-spherical clusters and present a unifying view of nine famous arbitrary-shaped clustering algorithms.We propose a strategy of extending AP in non-spherical clustering by constructing category similarity of objects.Leveraging the monotonicity that the clusters’number increases with the self-similarity in AP,we propose a model selection procedure that can determine the number of clusters adaptively.For the parameters introduced by extending AP in non-spherical clustering,we provide a grid-evolving strategy to optimize them automatically.The effectiveness of AdaSAP is evaluated by experiments on both synthetic datasets and real-world clustering tasks.Experimental results validate that the superiority of AdaSAP over benchmark algorithms like the classical AP and spectral clustering algorithms.展开更多
为使AP算法对图像进行聚类时充分考虑不同尺度的特征及有效利用未标记数据的特征,提出了结合特征金字塔网络的半监督AP聚类算法(Semi-supervised AP clustering Based on Feature Pyramid Networks, FPNSAP)。FPNSAP算法使用改进的特征...为使AP算法对图像进行聚类时充分考虑不同尺度的特征及有效利用未标记数据的特征,提出了结合特征金字塔网络的半监督AP聚类算法(Semi-supervised AP clustering Based on Feature Pyramid Networks, FPNSAP)。FPNSAP算法使用改进的特征金字塔网络来获得图像不同尺度的特征图,对不同大小的特征图进行融合,获得图像的高级语义特征,识别不同大小、不同实例的目标;k近邻标记更新策略可以动态增加标记数据集样本数量,充分利用未标记数据的特征,提高AP算法的聚类性能。FPNSAP算法与四个经典算法(FCH、SAP、DCN和DFCM)在Fashion-MNIST、YaleB和CIFAR-10数据集上进行实验对比,结果表明,FPNSAP算法具有较高的聚类性能,同时算法的鲁棒性更好。展开更多
文摘为解决密集Wi-Fi网络中重叠基本服务集(OBSS,overlapping basic service set)用户受到严重的同频干扰,导致接收信号信干噪比(SINR,signal to interference plus noise ratio)值低、数据传输速率低或者无法传输的问题,提出了一种基于改进遗传算法(IGA,improved genetic algorithm)选择接入点(AP,access point)协作集的多AP联合传输(JT,joint transmission)方案。首先,利用SINR阈值法将系统内所有站点(STA,station)分为中心用户和边缘用户,边缘用户采用JT方案。然后,以最大化边缘用户吞吐量为目标,利用IGA为每个边缘用户确定合适的协作AP形成其专属的AP协作集,最大化JT技术的潜在增益。仿真结果表明,所提方案在密集Wi-Fi网络场景下能够有效减小基本服务集(BSS,basic service set)间的同频干扰,提升BSS边缘用户的性能和吞吐量。
基金This work was supported by the National Natural Science Foundation of China(71771034,71901011,71971039)the Scientific and Technological Innovation Foundation of Dalian(2018J11CY009).
文摘Affinity propagation(AP)is a classic clustering algorithm.To improve the classical AP algorithms,we propose a clustering algorithm namely,adaptive spectral affinity propagation(AdaSAP).In particular,we discuss why AP is not suitable for non-spherical clusters and present a unifying view of nine famous arbitrary-shaped clustering algorithms.We propose a strategy of extending AP in non-spherical clustering by constructing category similarity of objects.Leveraging the monotonicity that the clusters’number increases with the self-similarity in AP,we propose a model selection procedure that can determine the number of clusters adaptively.For the parameters introduced by extending AP in non-spherical clustering,we provide a grid-evolving strategy to optimize them automatically.The effectiveness of AdaSAP is evaluated by experiments on both synthetic datasets and real-world clustering tasks.Experimental results validate that the superiority of AdaSAP over benchmark algorithms like the classical AP and spectral clustering algorithms.
文摘为使AP算法对图像进行聚类时充分考虑不同尺度的特征及有效利用未标记数据的特征,提出了结合特征金字塔网络的半监督AP聚类算法(Semi-supervised AP clustering Based on Feature Pyramid Networks, FPNSAP)。FPNSAP算法使用改进的特征金字塔网络来获得图像不同尺度的特征图,对不同大小的特征图进行融合,获得图像的高级语义特征,识别不同大小、不同实例的目标;k近邻标记更新策略可以动态增加标记数据集样本数量,充分利用未标记数据的特征,提高AP算法的聚类性能。FPNSAP算法与四个经典算法(FCH、SAP、DCN和DFCM)在Fashion-MNIST、YaleB和CIFAR-10数据集上进行实验对比,结果表明,FPNSAP算法具有较高的聚类性能,同时算法的鲁棒性更好。
