{"title":"可变步长修改的剪切LMS算法","authors":"Amin Aref, M. Lotfizad","doi":"10.1109/KBEI.2015.7436103","DOIUrl":null,"url":null,"abstract":"In this paper we introduce an Modified Clipped LMS (MCLMS) algorithm with a variable step size. In the MCLMS algorithm two parameters, the step size and the threshold control the convergence rate of the adaptive filter coefficients and also determine the final mean-square error. The computational complexity decreased dramatically by a large threshold. However, this selection results in a low convergence rate. Since the convergence time is inversely proportional to the step size, a large step size is often selected for fast convergence. But a large step size results in an increased final mean square error. Therefore in this paper we choose a large threshold and propose a variable step size for the MCLMS algorithm. The advantages of this proposed variable step size and a large threshold selection are that the computation complexity is low, final mean square error is low and that the convergence is fast.","PeriodicalId":168295,"journal":{"name":"2015 2nd International Conference on Knowledge-Based Engineering and Innovation (KBEI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":"{\"title\":\"Variable step size modified clipped LMS algorithm\",\"authors\":\"Amin Aref, M. Lotfizad\",\"doi\":\"10.1109/KBEI.2015.7436103\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this paper we introduce an Modified Clipped LMS (MCLMS) algorithm with a variable step size. In the MCLMS algorithm two parameters, the step size and the threshold control the convergence rate of the adaptive filter coefficients and also determine the final mean-square error. The computational complexity decreased dramatically by a large threshold. However, this selection results in a low convergence rate. Since the convergence time is inversely proportional to the step size, a large step size is often selected for fast convergence. But a large step size results in an increased final mean square error. Therefore in this paper we choose a large threshold and propose a variable step size for the MCLMS algorithm. The advantages of this proposed variable step size and a large threshold selection are that the computation complexity is low, final mean square error is low and that the convergence is fast.\",\"PeriodicalId\":168295,\"journal\":{\"name\":\"2015 2nd International Conference on Knowledge-Based Engineering and Innovation (KBEI)\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2015-11-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"5\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2015 2nd International Conference on Knowledge-Based Engineering and Innovation (KBEI)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/KBEI.2015.7436103\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 2nd International Conference on Knowledge-Based Engineering and Innovation (KBEI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/KBEI.2015.7436103","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
In this paper we introduce an Modified Clipped LMS (MCLMS) algorithm with a variable step size. In the MCLMS algorithm two parameters, the step size and the threshold control the convergence rate of the adaptive filter coefficients and also determine the final mean-square error. The computational complexity decreased dramatically by a large threshold. However, this selection results in a low convergence rate. Since the convergence time is inversely proportional to the step size, a large step size is often selected for fast convergence. But a large step size results in an increased final mean square error. Therefore in this paper we choose a large threshold and propose a variable step size for the MCLMS algorithm. The advantages of this proposed variable step size and a large threshold selection are that the computation complexity is low, final mean square error is low and that the convergence is fast.