Observation Leveraged Resampling-Free Particle Filter for Tracking of Rhythmic Biomedical Signals
Keywords:
Particle filtering, resampling, biomedical signals, electrocardiogram, root mean square error, computational timeAbstract
The particle filter is known to be a powerful tool to recursively estimate a hidden target state process using noisy observations from electronic sensor systems. The filter employs a set of particles that explore the state space using the Monte Carlo simulation of the target dynamics and then weighs them using the incoming observation. The congregation of the particles lead to probabilistic estimation of the true target state. However, the filter is effective only when the particles are drawn from regions of importance, i.e., the regions that contribute to the posterior probability density function. The traditional particle filter is known to suffer degeneracy as the target dynamics do not necessarily push the particles into regions of importance. This degeneracy problem can be overcome by either using a large number of particles or leveraging the incoming observation into the Monte Carlo sampling process. Since both solutions are not feasible, an additional resampling step was introduced to kill those particles that do not contribute to the posterior and replace them by copies of others that do. Furthermore, the recently proposed auxiliary particle filter and its variants improved upon the particle filter by mimicking the use of the incoming observation in the sampling process. However, the challenge of leveraging the incoming observation in the sampling process still remains a challenge. Moreover, these conventional filters still employ resampling which is a computationally expensive procedure. This paper proposes a novel particle filtering approach that takes into account the incoming observation into the sampling process without having to use resampling. This allows the particles to effectively explore the regions of importance and consequently result in fast and accurate filtering. The developed method is employed in tracking rhythmic biomedical signals and its accuracy and computational complexity are evaluated.
Downloads
References
Kim Sunghan, Lars Holmstrom, and James McNames, “Multiharmonic tracking using marginalized particle filters,” Proc. 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2008.
Kim Sunghan, Mateo Aboy, and James McNames. “Pulse pressure variation tracking using sequential monte carlo methods,” J. Biomedical Signal Processing and Control, Vol. 8, No. 4, pp. 333 – 340, 2013.
Kim Sunghan, Lars Andreas Holmstrom, and James McNames. “Tracking of rhythmical biomedical signals using the maximum a posteriori adaptive marginalized particle filter,” J. British Journal of Health Informatics and Monitoring, Vol. 2, No. 1, pp. 1–23, 2015.
Kim Sunghan, Fouzia Noor, Mateo Aboy, and James McNames, “A novel particle filtering method for estimation of pulse pressure variation during spontaneous breathing,” J. Biomedical engineering Online, Vol. 15, No. 1, pp. 1–18, 2016.
N. Gordon, David J. Salmond, and Adrian FM Smith, “Novel approach to nonlinear/non-Gaussian Bayesian state estimation,” In IEE proceedings F (radar and signal processing), vol. 140, no. 2, pp. 107–113. 1993.
M.S. Arulampalam, S. Maskell, N. Gordon, and T. Clapp, “A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking,” IEEE Trans. Signal Proc., vol 50, no. 2, pp. 174–188, 2002.
R. Douc, and O Cappe, “Comparison of resampling schemes for particle filtering,” In Proc. IEEE Symp. on Image and Signal Processing and Analysis, pp. 64–69, 2005.
J. D. Hol, Thomas B. Schon, and F. Gustafsson, “On resampling algorithms for particle filters,” Proc. 2006 IEEE Workshop on Nonlinear Statistical Signal Proc., pp. 79–82. 2006.
G. Kitagawa, “Monte Carlo filter and smoother for non-Gaussian nonlinear state space models,” J. of Computational and Graphical Statistics, pp. 1–25, 1996.
J. Liu, and R. Chen,” Sequential Monte Carlo methods for dynamic systems,” J. American Statistical Association, Vol. 93, No. 443, pp. 1032–1044, 1998.
P. B. Choppala, P. D. Teal, and M. R. Frean, “Resampling and Network Theory,” IEEE Trans. Signal and Information Processing over Networks, vol 08, pp. 106–119, 2022.
L. M. Murray, Anthony Lee, and Pierre E. Jacob, “Parallel resampling in the particle filter”, J. of Computational and Graphical Statistics, vol 25, no. 3, pp.789–805, 2016.
Mehdi Chitchian, Andrea Simonetto, Alexander S. van Amesfoort, and T. Keviczky, “Distributed Computation Particle Filters on GPU Architectures for Real-Time Control Applications,” IEEE Trans. Control Systems Technology, vol 21, no. 6, pp. 2224–2238, 2013.
A. Varsi, J Taylor, L Kekempanos, E. Knapp, and S. Maskell, “A Fast Parallel Particle Filter for Shared Memory Systems”, IEEE. Signal Proc. Letters, Vol. 27, pp. 1570–1574, 2020.
J.H. Kotecha, and P. M. Djuric, “Gaussian sum particle filtering,” IEEE Trans. Signal Processing, vol 51, No. 10, pp. 2602–2612, 2003.
M. Pitt, and Neil Shephard, “Filtering via simulation: Auxiliary particle filters,” J. American statistical Association, vol 94, no. 446, pp. 590–599, 1999.
V. Elvira, Luca Martino, Monica F. Bugallo, and Petar M. Djuric, “In search for improved auxiliary particle filters”, In Proc. IEEE European Signal Processing Conference (EUSIPCO), pp. 1637–1641, 2018.
Victor Elvira, Luca Martino, Monica Bugallo, and Petar M. Djuric, “Elucidating the auxiliary particle filter via multiple importance sampling”. IEEE Signal Processing Magazine, vol 36, no. 6, pp. 145–152, 2019.
J.P. Norton, and G. V. Veres, “Improvement of the particle filter by better choice of the predicted sample set,” Proc. of the IFAC, vol 35, no. 1, pp. 365–370, 2002.
M. Lin, Rong Chen, and Jun S. Liu, “Lookahead strategies for sequential Monte Carlo,” J. Statistical Science, vol 28, no. 1, pp.69–94, 2013.
Nicola Branchini, and Elvira Victor “, Optimized Auxiliary Particle Filters”, arXiv:2011.09317v1, 18 Nov 2020.
M. Rehman, S. C. Dass, and V. S. Asirvadam. “A weighted likelihood criteria for learning importance densities in particle filtering,” EURASIP Journal on Advances in Signal Proc., no. 1, pp. 1–19, 2018.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
