A Reconfigurable Data Flow Architecture for Signal Processing Applications

Hans Hauer,
Fraunhofer Institut Integrierte Schaltungen, Am Wolfsmantel 33, 91058 Erlangen, Germany
hauer@iis.fraunhofer.de

DOI: 10.36724/2664-066X-2021-7-5-2-6

SYNCHROINFO JOURNAL. Volume 7, Number 5 (2021). P. 2-6.

Abstract

This paper aims to devise a data flow model of computation for signal processing applications in which the operational nodes are signal/image processing functions such as Pixsum, Edge, Smooth [12]. These functions are configured during run time from a pool of reconfigurable FPGAS[4][5][6]. Thus because of the data flow model of computation, the signal processing functions execute concurrently. At the same time, these functions by exploiting their inherent spatial parallelism execute at high speed. There is a two fold speed up in the execution of image/signal processing applicationsone at the architecture level wherein a node of the dataflow model executes a digital signal processing (DSP) function rather than a low level machine operation. The second speed up is due to the fact that each DSP function is configured to execute in an FPGA by using maximally the concurrent operations that such a function permits. Another significant benefit that arises from our proposed architecture is that by reconfiguring an FPGA for a DSP function at run time, the reusability of the hardware elements results in reduced cost of operations. In this paper we provide an outline of the data flow architecture and its operational aspects.

KeywordsSemiconductor device modeling, RF IC Design, CMOS and BiCMOS circuit simulations, circuits for communications.

References

[1]. S.Y.Kung. VLSI Array Processor, PrenticeHall, 1988.
[2]. H.J. Siegal, et al. PASM: partitionable SIMD/MIMD system for image processing and pattern recognition, IEEE Trans. Computer., Vol.C30, no. 12, pp. 934-947, Dec 1981.
[3]. A.Sinha, P.C. Jain, V. Mitra. Asynchronous SIMD – A New Architecture For A Class of Image Processing Applications, proc. Fifth National Conf. On Communication, pp. 287-294, IIT Kharagpur, India. 1999.
[4]. Hungwen Li and Quentin F.Stout. Reconfigurable SIMD Massively parallel Computers, Proc. IEEE, Vol. 79, No.4, April 1991, pp. 429-443.
[5]. John Villasenor and William H.Mangione Smith. Configurable Computing, Scientific American Article, June 1997.
[6]. John Villasenor and Brad Hutchings. The Flexibility of Configurable Computing, IEEE Signal Processing Magazine,1998, pp. 67-83.
[7]. J.B. Dennis. Data Flow Supercomputers, Computer, pp. 48-56, Nov., 1980.
[8]. Shuichi Sakai, Yoshinori Yamaguchi, Kei Hiraki, Yeutsu Kodama, and Toshitsugu Yuba. An Architecture of a Dataflow Single Chip Processor, In Proc. 16th Annual Int’l Symposium on Computer Architecture, pp 46-53. ACM, 1989.
[9]. J.B. Dennis. MIT “Stream Data Types for signal processing”, Advanced topics in Dataflow Computing and Multithreading (Gao, Bic, Gaudiot eds.) IEEE Computer Society Press, 1995.
[10]. P. Netezki. Exploiting Data Parallelism in Signal Processing on a Dataflow Machine, ACM Annual Int’l Symposium on Computer Architecture, pp. 262-272, 1989.
[11]. Ben Lee and A.R. Hurson. Dataflow architectures and multithreading, IEEE Computer, pp. 27-28, Aug 1994. [12]. Edward R. Dougherty & Charles R. Giardina. Matrix Structured Image Processing, Prentice Hall, 1987.