Optimization of Digital Filter Structures Using Simulation Based Output Noise
-
- FUKUDA Satoshi
- Department of Electrical Engineering, Kanagawa University
-
- TOYOSHIMA Hisamichi
- Department of Electrical Engineering, Kanagawa University
Bibliographic Information
- Other Title
-
- シミュレーションに基づく誤差を用いたディジタルフィルタ構成の最適化
Search this article
Description
In the cascade realization of fixed-point digital filters, the output noise depends on the pole-zero pairing and the order of the second-order sections. The traditional optimization method is performed by analytically obtained scaling coefficients and roundoff errors. Therefore, that depends on the input signals and the structures of digital filters. In this paper, we propose a novel optimization method of filter structures by using output noise obtained from the discrete-time simulation. As an optimization algorhism, we employ the Genetic Algorithm (GA) and the fitness value is evaluated by the roundoff errors and overflow errors. This procedure can optimize under the complex criteria containing the scaling coefficient, roundoff errors, and overflow errors with no dependency of the input signals and the filter structures.
Journal
-
- IEICE technical report. Circuits and systems
-
IEICE technical report. Circuits and systems 98 (647), 57-62, 1999-03-09
The Institute of Electronics, Information and Communication Engineers
- Tweet
Details 詳細情報について
-
- CRID
- 1572543027236472832
-
- NII Article ID
- 110003198468
-
- NII Book ID
- AN10013094
-
- Text Lang
- ja
-
- Data Source
-
- CiNii Articles
