{"title":"非线性非递归不动点系统的自动SQNR确定","authors":"D. Ménard, R. Rocher, P. Scalart, O. Sentieys","doi":"10.5281/ZENODO.38366","DOIUrl":null,"url":null,"abstract":"Most of the digital signal processing applications are implemented in embedded systems which are based on fixed-point arithmetic. The reduction of the time-to-market requires the automation of the fixed-point specification determination. The accuracy evaluation is one of the most important stage of this process. In this paper, a new methodology for evaluating the quality of non-recursive and non-linear systems is presented. The fixed-point specification accuracy is automatically determined through the computation of the Signal-to-Quantization-Noise-Ratio (SQNR) expression. The theoretical approach used for computing the output noise power is detailed and the methodology developed for automating the accuracy evaluation is presented. Then, the quality of our estimation is evaluated through different experiments.","PeriodicalId":347658,"journal":{"name":"2004 12th European Signal Processing Conference","volume":"54 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2004-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"27","resultStr":"{\"title\":\"Automatic SQNR determination in non-linear and non-recursive fixed-point systems\",\"authors\":\"D. Ménard, R. Rocher, P. Scalart, O. Sentieys\",\"doi\":\"10.5281/ZENODO.38366\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Most of the digital signal processing applications are implemented in embedded systems which are based on fixed-point arithmetic. The reduction of the time-to-market requires the automation of the fixed-point specification determination. The accuracy evaluation is one of the most important stage of this process. In this paper, a new methodology for evaluating the quality of non-recursive and non-linear systems is presented. The fixed-point specification accuracy is automatically determined through the computation of the Signal-to-Quantization-Noise-Ratio (SQNR) expression. The theoretical approach used for computing the output noise power is detailed and the methodology developed for automating the accuracy evaluation is presented. Then, the quality of our estimation is evaluated through different experiments.\",\"PeriodicalId\":347658,\"journal\":{\"name\":\"2004 12th European Signal Processing Conference\",\"volume\":\"54 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2004-09-06\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"27\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2004 12th European Signal Processing Conference\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.5281/ZENODO.38366\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2004 12th European Signal Processing Conference","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.5281/ZENODO.38366","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Automatic SQNR determination in non-linear and non-recursive fixed-point systems
Most of the digital signal processing applications are implemented in embedded systems which are based on fixed-point arithmetic. The reduction of the time-to-market requires the automation of the fixed-point specification determination. The accuracy evaluation is one of the most important stage of this process. In this paper, a new methodology for evaluating the quality of non-recursive and non-linear systems is presented. The fixed-point specification accuracy is automatically determined through the computation of the Signal-to-Quantization-Noise-Ratio (SQNR) expression. The theoretical approach used for computing the output noise power is detailed and the methodology developed for automating the accuracy evaluation is presented. Then, the quality of our estimation is evaluated through different experiments.