A Parallel Architecture for the Self-Sorting FFT Algorithm

Univ Santiago, Fac Fis, Dept Electr & Comp, E 15706 Santiago, Spain and Univ Malaga, Fac Informat, Dept Arquitectura Comp, E 29013 Malaga, Spain
Journal of Parallel and Distributed Computing (Impact Factor: 1.18). 11/1995; 31(1):88–97. DOI: 10.1006/jpdc.1995.1147
Source: DBLP


The self-sorting (SS) algorithm is a highly efficient version of the fast Fourier transform (FFT), because, unlike the generally used algorithms, it does not require shuffling the sequence to be transformed (digit reversal), In this work, we propose a parallel architecture that implements the SS radix r (r ≥ 2) algorithm. The data flow of the algorithm is decomposed, in a natural way, into two sections that are implemented by means of FIFO queues located in the processors and an interprocessor connection network (perfect unshuffle). The resulting design is regular and modular, and, whenever possible, presents constant geometry. The total processing time required is nN/rQ cycles for a transform of size N = rn computed using Q = rq processors. Consequently, there are no cycle losses.

Full-text preview

Available from:
  • Source
    • "Most FFT algorithms require a presorted input block, or require some postsorting of the output block (bit-reversal or digit reversal). So-called self-sorting FFTs [10] avoid this, and can lead to particularly efficient vectorized FFT solutions. "
    [Show abstract] [Hide abstract]
    ABSTRACT: A major challenge of software-defined radio (SDR) is to realize many giga operations per second of flexible baseband processing within a power budget of only a few hundred mW. A heterogeneous hardware architecture with the programmable vector processor EVP as key component can support WLAN, UMTS, and other standards. A detailed rationale for the EVP architecture, based on the analysis of a number of key algorithms, as well as implementation and benchmarking results are described.
    Preview · Article · Jan 2005 · EURASIP journal on advances in signal processing
  • [Show abstract] [Hide abstract]
    ABSTRACT: A filter structure formed as a linear combination of stack filters (L-stack-filters), is studied. These types of filters include many useful filter classes, e.g., linear FIR filters and nonlinear threshold Boolean filters, and L- filters. An efficient algorithm for finding optimal filter coefficients under the mean squared error (MSE) criterion is derived. A subclass of the above filters, called FFT-ordered L-filters (FFT-LF), is studied in detail. In this case the bank of filters is formed according to a generalized structure of the FFT flowchart. It is shown that FFT-LFs effectively remove mixed Gaussian and impulsive noise. While possessing good characteristics of performance, FFT-LFs are simple in implementation. In the sense of implementation the most complicated FFT-LFs are the well-known L-filters. We suggest an efficient parallel architecture implementing FFT- LFs as well as a family of discrete orthogonal transforms including discrete Fourier and Walsh transforms. Both linear and nonlinear L-filter-type filters are implemented effectively on the architecture. Comparison with known architectures implementing both linear and nonlinear filter reveals advantages of the proposed architecture. An efficient implementation of L-stack-filters is also proposed.
    No preview · Article · Jan 1996 · Journal of Electronic Imaging
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we propose a software tool that generates single or multiple mode IFFT/FFT cores. This tool can be configured for different wireless communication standards that use OFDM modulation such as WLAN 802.11a/g, WMAN 802.16a, DAB, DVB-T. The resulting FFT/IFFT core fulfills the timing specifications of each standard with a minimum area.
    No preview · Conference Paper · Feb 2006
Show more