Open AccessParameter space dimension reduction of an adaptive interpolator during multidimensional signal differential compression
Author(s) -
Aleksey Maksimov,
М. В. Гашников,
Molodogvardejskaya street Photonics Ras
Publication year - 2019
Publication title -
ceur workshop proceedings
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.177
H-Index - 52
ISSN - 1613-0073
DOI - 10.18287/1613-0073-2019-2391-23-30
Subject(s) - interpolation (computer graphics) , dimension (graph theory) , algorithm , reduction (mathematics) , computer science , parameterized complexity , differential (mechanical device) , data compression , compression (physics) , multivariate interpolation , signal (programming language) , mathematics , computational complexity theory , mathematical optimization , artificial intelligence , computer vision , bilinear interpolation , motion (physics) , geometry , materials science , pure mathematics , engineering , composite material , programming language , aerospace engineering
We propose a new adaptive multidimensional signal interpolator for differential compression tasks. To increase the efficiency of interpolation, we optimize its parameters space by the minimum absolute interpolation error criterion. To reduce the complexity of interpolation optimization, we reduce the dimension of its parameter range. The correspondence between signal samples in a local neighbourhood is parameterized. Besides, we compare several methods for such parameterization. The developed adaptive interpolator is embedded in the differential compression method. Computational experiments on real multidimensional signals confirm that the use of the proposed interpolator can increase the compression ratio