Parallel implementation of linear repetitive processes identification using subspace algorithms
Dominik Kujawa
DOI: http://dx.doi.org/10.15439/2014F281
Citation: Position Papers of the 2014 Federated Conference on Computer Science and Information Systems, M. Ganzha, L. Maciaszek, M. Paprzycki (eds). ACSIS, Vol. 3, pages 79–83 (2014)
Abstract. This paper presents a new parallel approach to identification of linear repetitive processes based on subspace algorithms. Parallel realizations of these algorithms are tested on various graphic cards that use NVIDIA CUDA technology. The paper describes implementation of subspace identification algorithms and their parallel speedup, effciency, throughput, and delay. The parallel approach to the identification of linear repetitive processes based on subspace methods, presented in this paper, is very useful not only for time invariant LRPs but also for processes with dynamics that changes rapidly from pass to pass. A simulation example is included to illustrate the effectiveness of the proposed approach.