{"title":"A superconducting ternary systolic array processor","authors":"M. Morisue, Fu-Qiang Li","doi":"10.1109/ISMVL.1992.186772","DOIUrl":null,"url":null,"abstract":"A novel Josephson ternary systolic array processor for multiplication is proposed. The processor consists of two kinds of cells, one of which performs a partial multiplication and the other two functions of multiplication and addition simultaneously. The advantages of the processor are its very simple construction with a small number of SQUID gates and its very-high-speed operation and ultralow power dissipation. Information flows between cells in a pipeline fashion so that high performance can be achieved. The principle of the processor is described in detail, and simulation results for the multiplication of 2 trit*2 trit are presented. The results show that the pipeline operation can be executed in a cycle time of 450 ps.<<ETX>>","PeriodicalId":127091,"journal":{"name":"[1992] Proceedings The Twenty-Second International Symposium on Multiple-Valued Logic","volume":"12 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1992-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"[1992] Proceedings The Twenty-Second International Symposium on Multiple-Valued Logic","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISMVL.1992.186772","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
A novel Josephson ternary systolic array processor for multiplication is proposed. The processor consists of two kinds of cells, one of which performs a partial multiplication and the other two functions of multiplication and addition simultaneously. The advantages of the processor are its very simple construction with a small number of SQUID gates and its very-high-speed operation and ultralow power dissipation. Information flows between cells in a pipeline fashion so that high performance can be achieved. The principle of the processor is described in detail, and simulation results for the multiplication of 2 trit*2 trit are presented. The results show that the pipeline operation can be executed in a cycle time of 450 ps.<>