Fast, parallel implementation of particle filtering on the GPU architecture
Faculty of Information Technology, Pazmany Peter Catholic University, Prater str. 50/a, Budapest H-1083, Hungary
EURASIP Journal on Advances in Signal Processing, 2013:148, 2013
@article{gelencser2013fast,
title={Fast, parallel implementation of particle filtering on the GPU architecture},
author={Gelencs{‘e}r-Horv{‘a}th, Anna and Tornai, G{‘a}bor J{‘a}nos and Horv{‘a}th, Andr{‘a}s and Cserey, Gy{"o}rgy},
journal={EURASIP Journal on Advances in Signal Processing},
volume={2013},
number={1},
pages={148},
year={2013},
publisher={Springer}
}
In this paper, we introduce a modified cellular particle filter (CPF) which we mapped on a graphics processing unit (GPU) architecture. We developed this filter adaptation using a state-of-the art CPF technique. Mapping this filter realization on a highly parallel architecture entailed a shift in the logical representation of the particles. In this process, the original two-dimensional organization is reordered as a one-dimensional ring topology. We proposed a proof-of-concept measurement on two models with an NVIDIA Fermi architecture GPU. This design achieved a 411-micros kernel time per state and a 77-ms global running time for all states for 16,384 particles with a 256 neighbourhood size on a sequence of 24 states for a bearing-only tracking model. For a commonly used benchmark model at the same configuration, we achieved a 266-micros kernel time per state and a 124-ms global running time for all 100 states. Kernel time includes random number generation on the GPU as well as with curand. These results attest to the effective and fast use of the particle filter in high-dimensional, real-time applications.
September 28, 2013 by hgpu
