基于模糊贴近算法的海量异常激光数据挖掘方法研究
|
摘要
光纤网络中存在海量激光数据,对异常激光数据的挖掘存在较大困难。提出基于模糊贴近算法的海量异常激光数据挖掘方法,定义出一个激光数据处理域,在处理域中设置模糊贴近算法的模糊贴近度阈值,根据阈值定义正常隶属度函数与异常隶属度函数,得到正常、异常模糊数据集合。在激光数据库中,利用异常激光数据特征平均值表述数据动态特征,根据隶属度约束条件与网络参数实现模糊数据向精准数据的转化,通过分析数据动态特征完成对海量异常激光数据的定位预测。实验结果表明,所提方法的挖掘效率高,内存占用量小,数据聚合性好。
There is a large amount of laser data in the fiber optic network, which is difficult to excavate the abnormal laser data. The method of massive abnormal laser data mining based on fuzzy proximity algorithm is proposed. A laser data processing domain is defined. The threshold of the fuzzy proximity algorithm is set in the processing domain. Normal and abnormal membership functions are defined according to the threshold, and normal and abnormal fuzzy data sets are obtained. In laser database, using the average abnormal laser data characteristic expression data dynamic characteristics, according to the fuzzy membership degree constraint conditions and the network parameters implemente the transformation from fuzzy data to precise data. Through the analysis of dynamic characteristics, the data location prediction of massive abnormal laser data. The experimental results show that the method is very efficient, and the memory consumption is small and the data aggregation is good.
There is a large amount of laser data in the fiber optic network, which is difficult to excavate the abnormal laser data. The method of massive abnormal laser data mining based on fuzzy proximity algorithm is proposed. A laser data processing domain is defined. The threshold of the fuzzy proximity algorithm is set in the processing domain. Normal and abnormal membership functions are defined according to the threshold, and normal and abnormal fuzzy data sets are obtained. In laser database, using the average abnormal laser data characteristic expression data dynamic characteristics, according to the fuzzy membership degree constraint conditions and the network parameters implemente the transformation from fuzzy data to precise data. Through the analysis of dynamic characteristics, the data location prediction of massive abnormal laser data. The experimental results show that the method is very efficient, and the memory consumption is small and the data aggregation is good.
引文
[1] 叶琳琳,王晓曼,刘鹏.基于RMA的前视激光合成孔径雷达成像算法[J].半导体光电,2015,36(2):331-334.
[2] 彭正枫,杨晶,王元庆.基于TDC算法的激光内光路精细矫正[J].光电子技术,2015,35(2):106-110.
[3] Peláez V C,Ausín L,Ruiz M M,et al.Mining Fuzzy Association Rules Based on Parallel Particle Swarm Optimization Algorithm[J].Intelligent Automation & Soft Computing,2015,21(2):147-162.
[4] 孙韩,郝士琦,张岱,等.基于信道特征和中心极限定理的大气激光通信帧同步算法[J].光电技术应用,2016,31(5):12-16.
[5] 苗启广,郭雪,宋建锋,等.基于区域预测的LiDAR点云数据形态学滤波算法[J].激光与光电子学进展,2015,52(1):61-67.
[6] 刘忠宝,赵文娟.基于模糊大间隔最小球分类模型的恒星光谱离群数据挖掘方法[J].光谱学与光谱分析,2016,36(4):1245-1248.
[7] Sun S,Gong J,He J,et al.A spreading activation algorithm of spatial big data retrieval based on the spatial ontology model[J].Cluster Computing,2015,18(2):563-575.
[8] 任宏光,于海山,霍力君,等.激光引信云雾散射回波的近似算法与应用分析[J].电光与控制,2015,22(7):52-56.
[9] Isazadeh A,Mahan F,Pedrycz W.MFlexDT:multi flexible fuzzy decision tree for data stream classification[J].Soft Computing,2016,2(9):1-15.
[10] 吕峰,李丽娇,高云英,等.基于Hadoop在中医药数据挖掘中的应用[J].电子设计工程,2016,24(22):112-114.
[11] 韩瑞梅,杨晓,刘培.基于光学和雷达遥感影像融合的地类识别研究[J].红外技术,2015,37(11):949-956.
[12] Jin S,Cui W,Jin Z,et al.AF-DHNN:Fuzzy Clustering and Inference-Based Node Fault Diagnosis Method for Fire Detection.[J].Sensors,2015,15(7):17366-17396.
[13] Li W,Wang R,Song L,et al.Batch dynamically incremental c-means clustering algorithm based on rough fuzzy set[J].Journal of Computational Information Systems,2015,11(5):1553-1561.
[14] 李兴东,李满天,郭伟,等.TOF激光相机六自由度位姿变换估计[J].红外与激光工程,2015,44(7):2231-2238.
[15] 窦磊,郭大波,王晓凯.连续变量量子密钥分发多维数据协调算法优化[J].光学学报,2016,34(9):289-295.
[16] WU D,YAO J,LI H.Fault Diagnosis of Gear Based on the Variable Fuzzy Similarity Nearness[J].Open Mechanical Engineering Journal,2015,9(1):846-850.
[17] 肖胜.激光惯导系统流水式旋转矢量算法研究[J].光学与光电技术,2015,13(4):51-56.
[18] 修文正,申晋,肖莹莹,等.含噪动态光散射测量数据反演中正则化算法与Chahine算法的比较[J].光子学报,2016,45(11):153-159.
[19] 张新明,尹欣欣,涂强.动态迁移和椒盐变异融合生物地理学优化算法的高维多阈值分割[J].光学精密工程,2015,23(10):2943-2951.
[2] 彭正枫,杨晶,王元庆.基于TDC算法的激光内光路精细矫正[J].光电子技术,2015,35(2):106-110.
[3] Peláez V C,Ausín L,Ruiz M M,et al.Mining Fuzzy Association Rules Based on Parallel Particle Swarm Optimization Algorithm[J].Intelligent Automation & Soft Computing,2015,21(2):147-162.
[4] 孙韩,郝士琦,张岱,等.基于信道特征和中心极限定理的大气激光通信帧同步算法[J].光电技术应用,2016,31(5):12-16.
[5] 苗启广,郭雪,宋建锋,等.基于区域预测的LiDAR点云数据形态学滤波算法[J].激光与光电子学进展,2015,52(1):61-67.
[6] 刘忠宝,赵文娟.基于模糊大间隔最小球分类模型的恒星光谱离群数据挖掘方法[J].光谱学与光谱分析,2016,36(4):1245-1248.
[7] Sun S,Gong J,He J,et al.A spreading activation algorithm of spatial big data retrieval based on the spatial ontology model[J].Cluster Computing,2015,18(2):563-575.
[8] 任宏光,于海山,霍力君,等.激光引信云雾散射回波的近似算法与应用分析[J].电光与控制,2015,22(7):52-56.
[9] Isazadeh A,Mahan F,Pedrycz W.MFlexDT:multi flexible fuzzy decision tree for data stream classification[J].Soft Computing,2016,2(9):1-15.
[10] 吕峰,李丽娇,高云英,等.基于Hadoop在中医药数据挖掘中的应用[J].电子设计工程,2016,24(22):112-114.
[11] 韩瑞梅,杨晓,刘培.基于光学和雷达遥感影像融合的地类识别研究[J].红外技术,2015,37(11):949-956.
[12] Jin S,Cui W,Jin Z,et al.AF-DHNN:Fuzzy Clustering and Inference-Based Node Fault Diagnosis Method for Fire Detection.[J].Sensors,2015,15(7):17366-17396.
[13] Li W,Wang R,Song L,et al.Batch dynamically incremental c-means clustering algorithm based on rough fuzzy set[J].Journal of Computational Information Systems,2015,11(5):1553-1561.
[14] 李兴东,李满天,郭伟,等.TOF激光相机六自由度位姿变换估计[J].红外与激光工程,2015,44(7):2231-2238.
[15] 窦磊,郭大波,王晓凯.连续变量量子密钥分发多维数据协调算法优化[J].光学学报,2016,34(9):289-295.
[16] WU D,YAO J,LI H.Fault Diagnosis of Gear Based on the Variable Fuzzy Similarity Nearness[J].Open Mechanical Engineering Journal,2015,9(1):846-850.
[17] 肖胜.激光惯导系统流水式旋转矢量算法研究[J].光学与光电技术,2015,13(4):51-56.
[18] 修文正,申晋,肖莹莹,等.含噪动态光散射测量数据反演中正则化算法与Chahine算法的比较[J].光子学报,2016,45(11):153-159.
[19] 张新明,尹欣欣,涂强.动态迁移和椒盐变异融合生物地理学优化算法的高维多阈值分割[J].光学精密工程,2015,23(10):2943-2951.