Roundoff Noise Minimization Using Jointly Optimized Error Feedback and Realization for State-Space Digital Filters
| Autor: | Yoichi Hinamoto, Akimitsu Doi |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Iterative equation
Computer science MathematicsofComputing_NUMERICALANALYSIS 020206 networking & telecommunications Error feedback 02 engineering and technology Square matrix State space digital filters Norm (mathematics) Roundoff noise 0202 electrical engineering electronic engineering information engineering Minification Algorithm Scaling |
| Zdroj: | ICSIPA |
| DOI: | 10.1109/icsipa45851.2019.8977730 |
| Popis: | This paper deals with the subject of minimizing roundoff noise under the restrictions of $l_{2}$ -norm dynamic-range scaling for state space digital filters in which the subject is solved by simultaneously optimizing realization and error feedback. That is to say, a useful technique for minimizing roundoff noise is explored by relaxing the scaling restrictions into only one restriction on the trace of a square matrix and solving the relaxed subject with an efficient matrix iterative equation as well as two recursive scalar equations. The utility of the present technique is demonstrated through a numerical example. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|