Infall Syafalni, Rahmat Mulyawan, N. Sutisna, T. Adiono
{"title":"一种新的二项式展开式近似除法","authors":"Infall Syafalni, Rahmat Mulyawan, N. Sutisna, T. Adiono","doi":"10.1109/TSSA51342.2020.9310815","DOIUrl":null,"url":null,"abstract":"This paper presents a novel approximate divider circuit using binomial expansion. The circuit is approximated using the division between A and $L = {2^{\\left\\lfloor {{{\\log }_2}B} \\right\\rfloor }}$, where ⌊log2 B⌋ is the most significant bit value of the divisor B. After that, we use the sum of binomial coefficient to approximate the values. The approximate divider is much simpler and can be implemented using shift and add operations. Moreover, the complexity of the method is $\\mathcal{O}(n)$, where n is the number of bits. Experimental results show that the probability of errors is less than 0.18. The approximate circuit is useful for circuit applications contain rigorous and massive arithmetic operations such as artificial intelligence circuits.","PeriodicalId":166316,"journal":{"name":"2020 14th International Conference on Telecommunication Systems, Services, and Applications (TSSA","volume":"80 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A Novel Approximate Divider using Binomial Expansion\",\"authors\":\"Infall Syafalni, Rahmat Mulyawan, N. Sutisna, T. Adiono\",\"doi\":\"10.1109/TSSA51342.2020.9310815\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper presents a novel approximate divider circuit using binomial expansion. The circuit is approximated using the division between A and $L = {2^{\\\\left\\\\lfloor {{{\\\\log }_2}B} \\\\right\\\\rfloor }}$, where ⌊log2 B⌋ is the most significant bit value of the divisor B. After that, we use the sum of binomial coefficient to approximate the values. The approximate divider is much simpler and can be implemented using shift and add operations. Moreover, the complexity of the method is $\\\\mathcal{O}(n)$, where n is the number of bits. Experimental results show that the probability of errors is less than 0.18. The approximate circuit is useful for circuit applications contain rigorous and massive arithmetic operations such as artificial intelligence circuits.\",\"PeriodicalId\":166316,\"journal\":{\"name\":\"2020 14th International Conference on Telecommunication Systems, Services, and Applications (TSSA\",\"volume\":\"80 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-11-04\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2020 14th International Conference on Telecommunication Systems, Services, and Applications (TSSA\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/TSSA51342.2020.9310815\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 14th International Conference on Telecommunication Systems, Services, and Applications (TSSA","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/TSSA51342.2020.9310815","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
A Novel Approximate Divider using Binomial Expansion
This paper presents a novel approximate divider circuit using binomial expansion. The circuit is approximated using the division between A and $L = {2^{\left\lfloor {{{\log }_2}B} \right\rfloor }}$, where ⌊log2 B⌋ is the most significant bit value of the divisor B. After that, we use the sum of binomial coefficient to approximate the values. The approximate divider is much simpler and can be implemented using shift and add operations. Moreover, the complexity of the method is $\mathcal{O}(n)$, where n is the number of bits. Experimental results show that the probability of errors is less than 0.18. The approximate circuit is useful for circuit applications contain rigorous and massive arithmetic operations such as artificial intelligence circuits.