at40k-fft ATMEL Corporation, at40k-fft Datasheet

no-image

at40k-fft

Manufacturer Part Number
at40k-fft
Description
Fast Fourier Transform Intellectual Property Core At40k Fpgas
Manufacturer
ATMEL Corporation
Datasheet
Features
Description
The Fast Fourier Transform (FFT) processor is a FFT engine developed for the
AT40K family of Field Programmable Gate Arrays (FPGAs). The design is based on a
decimation-in-frequency radix-2 algorithm and employs in-place computation to opti-
mize memory usage. In order to operate the processor, data must first be loaded into
the internal RAM. The processor is then instructed to compute the FFT, overwriting
the input data in the RAM with the results. Upon completion of the FFT, the results
may be read out from the RAM via the output data port.
Decimation in frequency radix-2 FFT algorithm.
256-point transform.
12-bit fixed point arithmetic.
Fixed scaling to avoid numeric overflow.
Requires no external memory, i.e. uses on chip RAM and ROM.
External access to on-chip RAM for data IO.
Clock speed of 21 MHz.
98 ms processing time.
70% utilization of AT40K30 logic resources.
48% utilization of AT40K30 RAM resources.
AT40K FPGA
IP Core
AT40K-FFT
Rev. 1132A–08/98
1

Related parts for at40k-fft

at40k-fft Summary of contents

Page 1

... RAM. The processor is then instructed to compute the FFT, overwriting the input data in the RAM with the results. Upon completion of the FFT, the results may be read out from the RAM via the output data port. AT40K FPGA IP Core AT40K-FFT Rev. 1132A–08/98 1 ...

Page 2

... Figure 2. Example FFT processor configuration 1. Real input Imag input Input Clock Figure 3. Example FFT processor configuration 2. AT40K-FFT 2 Address Address Counter Counter Data RAM Butterfly ...

Page 3

... AT40K FPGA. These problems can be overcome if the butterfly calcula- tion is performed over two clock cycles. This balances the ...

Page 4

... T Re(B’)=(Im(-A+B)xIm(W )+Re(-A+B Re(W ))/2=Re((A-B)xW ). The other is fed into a differencing circuit which subtracts the two results, divides by two and truncates the result to 12 bits with rounding, yielding, AT40K-FFT 4 Delay Delay ...

Page 5

... MatLab FFT function. Timing Analysis Timing analysis of the design indicated a maximum clock speed of 21.2 MHz when using the AT40K30 part. Investi- gations into the limiting data path revealed this to be the delay from the twiddle factor address counter output, through the asynchronous twiddle factor ROM to the butter- fly’ ...

Page 6

... For large transform lengths the use of an external dual ported memory should be considered, this may also pro- vide faster data transfer times by reducing on chip bus loadings. AT40K-FFT 6 System performance could potentially be improved by pro- viding suitable buffering to permit concurrent data process- ing and IO. Double buffering designs would most probably have to use external memory devices ...

Page 7

7 ...

Page 8

... Marks bearing and/or are registered trademarks and trademarks of Atmel Corporation. Terms and product names in this document may be trademarks of others. Atmel Operations Atmel Colorado Springs 1150 E. Cheyenne Mtn. Blvd. Colorado Springs, CO 80906 ...

Related keywords