Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Data access method and device for parallel FFT (Fast Fourier Transform) computation

A data access and data technology, applied in computing, electrical digital data processing, memory systems, etc., can solve the problems of complex read and write address calculations, and difficulty in implementing parallel FFT algorithms.

Active Publication Date: 2012-04-11
BEIJING SMART LOGIC TECH CO LTD
View PDF6 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This patent requires a dedicated memory, data buffer and selector, and the calculation of read and write addresses is complex, making it difficult to implement parallel FFT algorithms with different data lengths and different read and write granularities

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Data access method and device for parallel FFT (Fast Fourier Transform) computation
  • Data access method and device for parallel FFT (Fast Fourier Transform) computation
  • Data access method and device for parallel FFT (Fast Fourier Transform) computation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be described in further detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0037] The following symbols are agreed upon during the description:

[0038] ■N: FFT data length, must be a power of 2;

[0039] ■B: Parallel granularity bit width, 2 B is the parallel granularity;

[0040] ■G: data group index value bit width, G=log 2 N-B;

[0041] ■g: data group index value;

[0042] ■b: index value within the group.

[0043] In order to realize the read, write and calculation of parallel FFT, first define the parallel granularity 2 B , (B is a positive integer), the parallel granularity refers to the number of data that can be read in parallel from the memory and independently perform butterfly calculations. like Figure 4 As shown in the figure, it is assumed that the FFT data length N=64, an...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a data access method and device for parallel FFT (Fast Fourier Transform) computation. The data access method comprises the following steps of: storing FFT computation data and rotating factors in a multi-granularity parallel memory, and grouping the FFT data and the corresponding rotating factors in the whole computation process by using a united butterfly pattern expression, wherein data of each subgroup contain a plurality of butterfly patterns and can be subjected to parallel computation; and simultaneously, generating a data address and a rotating factor coefficient address of each subgroup conveniently according to the butterfly pattern expression, and reading and writing the data and corresponding rotating factors in parallel from the multi-granularity memory according to different reading and writing granularities. The invention also provides a data access device for parallel FFT computation. By utilizing the data access method and device disclosed by the invention, memory conflict cannot be caused in the data reading and writing process, and no extra step is needed for sequencing the read and written data; meanwhile, according to the invention, parallel granularities can be appointed flexibly as specific conditions.

Description

technical field [0001] The invention mainly relates to the field of data access, in particular to a data access method and device for parallel FFT calculation. Background technique [0002] Signal processing systems often need to convert signal content in the time domain and frequency domain, and the Fast Fourier Transform algorithm (FFT) can perform signal conversion between the time domain and the frequency domain. Compared with other conversion algorithms, the fast Fourier transform algorithm has the advantages of uniform structure and less calculation, so it is widely used in signal processing systems. [0003] The FFT algorithm inputs N data and outputs N data; it is generally called the forward transform from the time domain to the frequency domain, and the inverse transform from the frequency domain to the time domain. There are many ways to implement the FFT algorithm, but they are all evolved from the Cooley-Tukey algorithm. For N data points, the Cooley-Tukey alg...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F9/302G06F17/14
Inventor 谢少林林啸蒿杰薛晓军汪涛尹磊祖
Owner BEIJING SMART LOGIC TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com