一种基于小波处理的SVM数据分类算法  

A SVM Data Classification Algorithm Based on Wavelet Processing

作　　者：汪思冒 陆格格 郝广凯 魏祎 雷鸣 WANG Simao;LU Gege;Hao Guangkai;WEI Yi;LEI Ming(Shanghai Institute of Aerospace Electronics Technology,Shanghai 201109,China)

机构地区：[1]上海航天电子技术研究所,上海201109

出　　处：《通信电源技术》2023年第15期14-17,共4页Telecom Power Technology

摘　　要：文章研究了一种基于小波处理的支持向量机(Support Vector Machine,SVM)分类算法,该算法能够通过小波分析对输入信号进行去噪和特征提取,使用SVM进行机械学习,并对目标信号进行精准分类。同时,该算法能够实现自动计算和高精度分类,完成难以通过人工手段实施的小信号精准判别。此外,经过参数的调整选取,该算法具有强鲁棒性与高灵敏度,可良好应用于微小信号识别中,大大提高分类准确性,减少分类所需时间。This paper studies a Support Vector Machine(SVM)classification algorithm based on wavelet processing,which can denoise and extract features of input signals through wavelet analysis,use SVM for Rote learning,and accurately classify target signals.At the same time,this algorithm can achieve automatic calculation and high-precision classification,completing precise discrimination of small signals that are difficult to implement through manual means.In addition,after parameter adjustment and selection,the algorithm has strong robustness and high sensitivity,which can be well applied in small signal recognition,greatly improving classification accuracy and reducing classification time.

关 键 词：小波处理 支持向量机(SVM) 分类算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]