{"title":"VLSI architecture of a scalable matrix transposer","authors":"O. Fatemi, S. Panchanathan","doi":"10.1109/ICISS.1996.552445","DOIUrl":null,"url":null,"abstract":"In this paper, we present an ASIC implementation of matrix transposition (MT) using FPGA. MT is an important operation in 2-dimensional signal and image processing applications. Recently several parallel and pipelined architectures for real-time implementation of MT have been presented in the literature. However these implementations are not scalable. For example an architecture for the transposition of an 8/spl times/8 matrix cannot be directly obtained from a 4/spl times/4 implementation. We propose a parallel and pipelined architecture for real-time MT. This architecture is modular and cascadable. In addition it has a small execution time and low communication complexity.","PeriodicalId":131620,"journal":{"name":"1996 Proceedings. Eighth Annual IEEE International Conference on Innovative Systems in Silicon","volume":"56 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1996-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"1996 Proceedings. Eighth Annual IEEE International Conference on Innovative Systems in Silicon","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICISS.1996.552445","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
In this paper, we present an ASIC implementation of matrix transposition (MT) using FPGA. MT is an important operation in 2-dimensional signal and image processing applications. Recently several parallel and pipelined architectures for real-time implementation of MT have been presented in the literature. However these implementations are not scalable. For example an architecture for the transposition of an 8/spl times/8 matrix cannot be directly obtained from a 4/spl times/4 implementation. We propose a parallel and pipelined architecture for real-time MT. This architecture is modular and cascadable. In addition it has a small execution time and low communication complexity.