Unlock instant, AI-driven research and patent intelligence for your innovation.

A Block Floating Point Method for FFT Processor

A block floating point, processor technology, applied in the direction of electrical digital data processing, digital data processing components, instruments, etc., can solve the problem of being unable to support FFT processors with adjustable data block sizes

Inactive Publication Date: 2019-09-17
BEIJING INSTITUTE OF TECHNOLOGYGY
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a block floating-point method of FFT processor, to solve the problem that the FFT processor with adjustable data block size cannot be supported in the prior art

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
  • A Block Floating Point Method for FFT Processor
  • A Block Floating Point Method for FFT Processor
  • A Block Floating Point Method for FFT Processor

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0058] Assume that the maximum data parallelism P that the FFT processor can access in parallel is 16. And the FFT processor uses the mixed base algorithm to decompose the N-point FFT operation into m-level butterfly operations:

[0059]

[0060] and have:

[0061]

[0062] According to the data conflict-free access algorithm, the formula of the heap where the data block is located in the data memory is:

[0063]

[0064] in, for n 1 in reverse order.

[0065] At stage i, the FFT processor performs a radix-N i operation, if N i =16, the size of the data block is 16, and the parameter n exists in the same data block i There are 16 different values; if N i ≠16 and i≠1, the data block size is

[0066]

[0067] in for right Round down. parameter n exists in the same data block i have N i different values ​​and parameter n' 1 have different values, where n' 1 for n 1 the low bit, n' 1 The corresponding number of possible values ​​is no i The c...

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 relates to the technical field of FFT (fast Fourier transform) processors, and discloses a block-floating-point method for a FFT processor. The block-floating-point method includes the steps: performing corresponding operations for data blocks of the level in some level of the FFT processor; performing corresponding zooming operations for the data blocks according to the operation results, calculating out block-floating-point indexes of the data blocks of the level, and storing the block-floating-point indexes according to accessing formulas of the block-floating-point indexes; reading the block-floating-point indexes according to the accessing formulas of the block-floating-point indexes in nest level of some level of the FFT processor, and performing corresponding aligned operations according to the block-floating-point indexes. The block-floating-point method has the advantage that sizes of the data blocks can be properly adjusted as required.

Description

technical field [0001] The invention relates to the technical field of FFT processors, in particular to a block floating-point method of an FFT processor. Background technique [0002] FFT is a fast implementation of the discrete Fourier transform (DFT), which converts data between the time and frequency domains. In a storage-based fixed-point FFT processor, due to the limited word length of the data in the memory, the calculation results of the arithmetic unit usually need to be scaled before being stored back into the memory. However, this will cause precision loss to the calculation results, that is, FFT The more stages required for the operation and the larger the base, the greater the loss of the accuracy of the final calculation result. [0003] Based on this consideration, it is necessary to use block floating-point technology to solve the precision loss caused by scaling. [0004] In existing block floating-point techniques for memory-based FFT processors, the size...

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 Patents(China)
IPC IPC(8): G06F17/14G06F7/57
Inventor 刘大可刘劭晗
Owner BEIJING INSTITUTE OF TECHNOLOGYGY