虚拟化云计算动态移动数据实时去噪处理系统设计

2018-11-13谢芳

现代电子技术 2018年22期

谢芳

摘要：现有系统进行虚拟化云计算动态移动数据去噪时，存在数据单个体运行且一次性完成去噪处理的问题。为提高云计算结果的准确性，设计新的虚拟化云计算动态移动数据实时去噪处理系统。该设计方法主要分为两个层次：基于Hadoop云计算平台描述系统硬件设计的三项功能，结合三项功能从前端控制层、运行层、用户层对系统C/S模式进行设计，实现海量动态移动数据的并行化处理；系统软件通过AFLS、并行处理中间件、查询服务器、DBMS、OTS五大结构，实现云计算动态移动数据的查询和简单去噪处理，采用网闸实时去噪处理方法对云计算动态移动数据进行二次实时去噪处理。实验结果表明，该系统与Matlab小波去噪系统和FPGA去噪系统相比，最高鲁棒性分别提高0.02%和0.08%，最低鲁棒性分别提高0.03%和0.05%；相同噪声数量下，去噪误差率最大值优于其他两种方法，分别为0.24%，0.29%；所设计的系统弱化了现有方法的不足，具有去噪精度高、稳定性好的优势。

关键词：云计算；动态移动数据；去噪处理； C/S模式；系统设计；鲁棒性

中图分类号： TN929.5?34； TP314 文献标识码： A 文章编号： 1004?373X（2018）22?0017?04

Abstract： The existing system has the problems of single data running and one?time denoising processing during the denoising of virtualized cloud computing dynamic mobile data. Therefore， a new real?time denoising processing system for virtualized cloud computing dynamic mobile data is designed to improve the accuracy of cloud computing results. The design method is mainly divided into two levels. Three functions of system hardware design based on the Hadoop cloud computing platform are described. Combining with the three functions， the C/S mode of the system consisting of the front?end control layer， running layer and user layer is designed to realize parallelization of massive dynamic mobile data. In the system software， five structures of the AFLS， parallel processing middleware， query server， DBMS and OTS are used to realize query and simple denoising processing of cloud computing dynamic mobile data. The GAP real?time denoising processing method is adopted to conduct secondary real?time denoising processing of cloud computing dynamic mobile data. The experimental results show that， the maximum robustness of the designed system is 0.02% and 0.08% higher respectively than that of the Matlab wavelet denoising system and FPGA denoising system， and the minimum robustness of the designed system is 0.03% and 0.05% higher respectively than that of the Matlab wavelet denoising system and FPGA denoising system； under the same quantity of noises， the maximum value of the denoising error rate is 0.24% and 0.29% lower than that of the other two methods； the designed system attenuates the shortcomings of the existing methods and has the advantages of high denoising precision and good stability.

Keywords： cloud computing； dynamic mobile data； denoising processing； C/S mode； system design； robustness