Design of 2-D FIR Filters by a Feedback Neural Network

  • Authors:

  • D. Bhattacharya;A. Antoniou

  • Affiliations:

  • Lucent Technologies, 1247 S Cedar Crest Blvd., Allentown, PA 18103, USA;Department of Electrical and Computer Engineering University of Victoria, P.O.Box 3055, Victoria, B.C., Canada V8W 3P6

  • Venue:
  • Multidimensional Systems and Signal Processing
  • Year:
  • 1999

Quantified Score

Hi-index 0.00

Visualization

Abstract

AHopfield-type neural network for the design of 2-D FIR filtersis proposed. The network is contrived to have an energy functionthat coincides with the sum-squared error of the approximationproblem at hand and by ensuring that the energy is a monotonicdecreasing function of time, the approximation problem can besolved. Two solutions are obtained. In the first the 2-D FIRfilter is designed on the basis of a specified amplitude responseand in the second a filter that has specified maximum passbandand stopband errors is designed. The network has been simulatedwith HSPICE and design examples are included to show that thisis an efficient way of solving the approximation problem for2-D FIR filters. The neural network has high potential for implementationin analog VLSI and can, as a consequence, be used in real-timeapplications.