{"title":"基于块浮点缩放技术的DVB-T 8k点流水线FFT/IFFT压缩存储","authors":"Hui-Gon Kim, Ki-Tae Yoon, Jin-Sun Youn, J. Choi","doi":"10.1109/ISWPC.2009.4800549","DOIUrl":null,"url":null,"abstract":"We Propose a 2K/4K/8K point FFT (Fast Fourier Transform) for OFDM (Orthogonal Frequency Division Multiplexing) of DVB-H (Digital Video Broadcast Terrestrial) Receiver. The proposed FFT architecture utilizes cascaded radix-4 single path feedback (SDF) structure based on the Radix-2/Radix-4 FFT algorithm.[11] We use block floating point scaling technique in order to increase SQNR. The 2K/8K FFT consists of 5 cascaded stages of radix-4 and 3 stages of radix-2 butterfly units. The SQNR of 58dB is achieved with 10-bit data input, 14-bit internal data and twiddle factors, and 18-bit data output. The core has 75,804 gates with 204,672 bits of RAM and 33,572 bits of ROM using 0.18um CMOS technology","PeriodicalId":383593,"journal":{"name":"2009 4th International Symposium on Wireless Pervasive Computing","volume":"4 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2009-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"8K-point Pipelined FFT/IFFT with Compact Memory for DVB-T using Block Floating-point Scaling Technique\",\"authors\":\"Hui-Gon Kim, Ki-Tae Yoon, Jin-Sun Youn, J. Choi\",\"doi\":\"10.1109/ISWPC.2009.4800549\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We Propose a 2K/4K/8K point FFT (Fast Fourier Transform) for OFDM (Orthogonal Frequency Division Multiplexing) of DVB-H (Digital Video Broadcast Terrestrial) Receiver. The proposed FFT architecture utilizes cascaded radix-4 single path feedback (SDF) structure based on the Radix-2/Radix-4 FFT algorithm.[11] We use block floating point scaling technique in order to increase SQNR. The 2K/8K FFT consists of 5 cascaded stages of radix-4 and 3 stages of radix-2 butterfly units. The SQNR of 58dB is achieved with 10-bit data input, 14-bit internal data and twiddle factors, and 18-bit data output. The core has 75,804 gates with 204,672 bits of RAM and 33,572 bits of ROM using 0.18um CMOS technology\",\"PeriodicalId\":383593,\"journal\":{\"name\":\"2009 4th International Symposium on Wireless Pervasive Computing\",\"volume\":\"4 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2009-02-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2009 4th International Symposium on Wireless Pervasive Computing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ISWPC.2009.4800549\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2009 4th International Symposium on Wireless Pervasive Computing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISWPC.2009.4800549","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
8K-point Pipelined FFT/IFFT with Compact Memory for DVB-T using Block Floating-point Scaling Technique
We Propose a 2K/4K/8K point FFT (Fast Fourier Transform) for OFDM (Orthogonal Frequency Division Multiplexing) of DVB-H (Digital Video Broadcast Terrestrial) Receiver. The proposed FFT architecture utilizes cascaded radix-4 single path feedback (SDF) structure based on the Radix-2/Radix-4 FFT algorithm.[11] We use block floating point scaling technique in order to increase SQNR. The 2K/8K FFT consists of 5 cascaded stages of radix-4 and 3 stages of radix-2 butterfly units. The SQNR of 58dB is achieved with 10-bit data input, 14-bit internal data and twiddle factors, and 18-bit data output. The core has 75,804 gates with 204,672 bits of RAM and 33,572 bits of ROM using 0.18um CMOS technology