面向监测数据压缩的自适应SDT算法  

Adaptive SDT algorithm for monitoring data compression

作　　者：张宗华[1] 叶志佳[2] 牛新征[2] ZHANG Zonghua;YE Zhijia;NIU Xinzheng(Ministry of Information and Communication,Beijing Electric Power Hospital,State Grid Corporation of China,Beijing 100073,China;School of Computer Science and Engineering,University of Electronic Science and Technology of China,Chengdu 611731,China)

机构地区：[1]国家电网公司北京电力医院信息通讯部,北京100073 [2]电子科技大学计算机科学与工程学院,四川成都611731

出　　处：《中国测试》2017年第2期104-108,共5页China Measurement & Test

基　　金：国家自然科学基金项目(61300192);中央高校基本科研业务费项目(ZYGX2014J052);北京电力医院一体化运维监控与管理项目(HW2015000759)

摘　　要：为降低IT运维系统的实时监测数据量、提高数据存储效率,提出一种自适应的旋转门算法(adaptive swinging door trending,ASDT)。针对传统SDT算法存在抗噪性弱、参数选取困难等缺陷,ASDT首先通过最小二乘平滑处理,减小噪声数据对SDT趋势判断的影响;然后通过改进死区限值过滤算法,对经平滑处理后的数据进行压缩;最后基于相邻压缩区间标准差变化,自适应调整压缩精度参数。实验结果表明:在保证数据保真度的前提下,ASDT的仿真数据和真实数据上的压缩比分别提高60%和24%以上。To reduce the amount of monitoring data of IT operation and maintenance system and improve the efficiency of data storage, an adaptive SDT algorithm named adaptive swinging door trending (ASDT) was proposed. To address problems such as the weak resistance to noise and the difficulty of parameter selection of traditional SDT algorithm, ASDT firstly adopts least-squares to smooth the original data to reduce the influence of noise to SDT trend judgment; then, it combines with improved boxcar-back slope algorithm to compress the data after smoothing; finally, it adjusts the parameters of compression accuracy adaptively based on the changes of standard deviation of adjacent interval. Results of experiments conducted on the simulation data and real data show that on the premise of guaranteeing the data fidelity, ASDT’s compression ratio is increased by over 60% and 24% respectively.

关 键 词：数据压缩 旋转门算法 平滑处理 自适应调整 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]