National Technical University of Athens (NTUA)
Θεωρητική και πρακτική ανάλυση αριθμητικής συμπεριφοράς σημαντικών αλγορίθμων για ψηφιακή επεξεργασία σημάτων
Abstract
dc:descriptionThe doctorial thesis is part of the research into the study of the error which results from the finite numerical precision in the arithmetic floating-point in recursive computational algorithms. In this paper an original method for estimating the accumulated numerical error is introduced, which is particularly efficient since it provides us with significant precision, it is implemented independently from the statistical characteristics of the input signals and it is applicable eeven in more complex algorithms. This new method allows the accurate knowledge in the number of erroneous digits with which all the quantities under algorithm control are computed in every execution of the algorithm, Specific examples of algorithms in which the above mentioned are implemented , are the algorithms Kalman, Fast Kalman, the algorithms for F.I.R. and I.I.R. filtering, the Matrix inversion algorithms with classical methods e.t.c.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2001
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Ioannidou, Persefoni
- Ιωαννίδου, Περσεφόνη
Subjects
dc:subject × 12- Αλγόριθμοι
- Αναδρομικός αλγόριθμος
- Αριθμητικό σφάλμα
- Αριθμητική κινητής υποδιαστολής
- Algorithms
- Recursive algorithms
- Numerical error
- Arithmetic floating point
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/12712