{"title":"Systolic structures for second-order recursive filters","authors":"S. Sunder, P. Diniz, F. El-Guibaly, A. Antoniou","doi":"10.1109/PACRIM.1991.160789","DOIUrl":null,"url":null,"abstract":"Two methods for the realization of second-order digital filters in terms of systolic structures are proposed, which yield a variety of structures. In some the input is pipelined, in others the output is pipelined, and in one case all the signals are pipelined. The structures are always modular and the designer can choose a structure with a larger number of simple processing elements (PEs) or one with a small number of more complex PEs. The first method yields simple and economical structures. In applications where limit cycles or quantization noise can present a problem, one of the more complicated structures obtained by the second method may be used.<<ETX>>","PeriodicalId":289986,"journal":{"name":"[1991] IEEE Pacific Rim Conference on Communications, Computers and Signal Processing Conference Proceedings","volume":"7 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1991-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"[1991] IEEE Pacific Rim Conference on Communications, Computers and Signal Processing Conference Proceedings","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/PACRIM.1991.160789","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Two methods for the realization of second-order digital filters in terms of systolic structures are proposed, which yield a variety of structures. In some the input is pipelined, in others the output is pipelined, and in one case all the signals are pipelined. The structures are always modular and the designer can choose a structure with a larger number of simple processing elements (PEs) or one with a small number of more complex PEs. The first method yields simple and economical structures. In applications where limit cycles or quantization noise can present a problem, one of the more complicated structures obtained by the second method may be used.<>