[Zurück]

O. Sluciak, H. Straková, M. Rupp, W. Gansterer:
"Distributed Gram-Schmidt orthogonalization with simultaneous elements refinement";
EURASIP Journal on Advances in Signal Processing, 2016:25 (2016), 25; S. 1 - 13.

We present a novel distributed QR factorization algorithm for orthogonalizing a set of vectors in a decentralized
wireless sensor network. The algorithm is based on the classical Gram-Schmidt orthogonalization with all projections
and inner products reformulated in a recursive manner. In contrast to existing distributed orthogonalization
algorithms, all elements of the resulting matrices Q and R are computed simultaneously and refined iteratively after
each transmission. Thus, the algorithm allows a trade-off between run time and accuracy. Moreover, the number of
transmitted messages is considerably smaller in comparison to state-of-the-art algorithms. We thoroughly study its
numerical properties and performance from various aspects. We also investigate the algorithm´s robustness to link
failures and provide a comparison with existing distributed QR factorization algorithms in terms of communication
cost and memory requirements.

We present a novel distributed QR factorization algorithm for orthogonalizing a set of vectors in a decentralized
wireless sensor network. The algorithm is based on the classical Gram-Schmidt orthogonalization with all projections
and inner products reformulated in a recursive manner. In contrast to existing distributed orthogonalization
algorithms, all elements of the resulting matrices Q and R are computed simultaneously and refined iteratively after
each transmission. Thus, the algorithm allows a trade-off between run time and accuracy. Moreover, the number of
transmitted messages is considerably smaller in comparison to state-of-the-art algorithms. We thoroughly study its
numerical properties and performance from various aspects. We also investigate the algorithm´s robustness to link
failures and provide a comparison with existing distributed QR factorization algorithms in terms of communication
cost and memory requirements.

Distributed processing, Gram-Schmidt orthogonalization, QR factorization

http://dx.doi.org/10.1186/s13634-016-0322-6

http://publik.tuwien.ac.at/files/PubDat_248334.pdf

Projektleitung Markus Rupp:
Signal and Information Processing in Science and Engineering II: Theory and Implementation of Distributed Algorithms