TY - GEN
T1 - A new parallel implementation for particle filters and its application to adaptive waveform design
AU - Miao, Lifeng
AU - Zhang, Jun Jason
AU - Chakrabarti, Chaitali
AU - Papandreou-Suppappola, Antonia
PY - 2010/12/27
Y1 - 2010/12/27
N2 - Sequential Monte Carlo particle filters (PFs) are useful for estimating nonlinear non-Gaussian dynamic system parameters. As these algorithms are recursive, their real-time implementation can be computationally complex. In this paper, we analyze the bottlenecks in existing parallel PF algorithms, and we propose a new approach that integrates parallel PFs with independent Metropolis-Hastings (PPF-IMH) algorithms to improve root mean-squared estimation error performance. We implement the new PPF-IMH algorithm on a Xilinx Virtex-5 field programmable gate array (FPGA) platform. For a one-dimensional problem and using 1,000 particles, the PPF-IMH architecture with four processing elements utilizes less than 5% Virtex-5 FPGA resources and takes 5.85 μs for one iteration. The algorithm performance is also demonstrated when designing the waveform for an agile sensing application.
AB - Sequential Monte Carlo particle filters (PFs) are useful for estimating nonlinear non-Gaussian dynamic system parameters. As these algorithms are recursive, their real-time implementation can be computationally complex. In this paper, we analyze the bottlenecks in existing parallel PF algorithms, and we propose a new approach that integrates parallel PFs with independent Metropolis-Hastings (PPF-IMH) algorithms to improve root mean-squared estimation error performance. We implement the new PPF-IMH algorithm on a Xilinx Virtex-5 field programmable gate array (FPGA) platform. For a one-dimensional problem and using 1,000 particles, the PPF-IMH architecture with four processing elements utilizes less than 5% Virtex-5 FPGA resources and takes 5.85 μs for one iteration. The algorithm performance is also demonstrated when designing the waveform for an agile sensing application.
KW - FPGA
KW - Independent Metropolis-Hastings algorithm
KW - Parallel architecture
KW - Particle filter
UR - http://www.scopus.com/inward/record.url?scp=78650356474&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=78650356474&partnerID=8YFLogxK
U2 - 10.1109/SIPS.2010.5624820
DO - 10.1109/SIPS.2010.5624820
M3 - Conference contribution
AN - SCOPUS:78650356474
SN - 9781424489336
T3 - IEEE Workshop on Signal Processing Systems, SiPS: Design and Implementation
SP - 19
EP - 24
BT - 2010 IEEE Workshop on Signal Processing Systems, SiPS 2010 - Proceedings
T2 - 2010 IEEE Workshop on Signal Processing Systems, SiPS 2010
Y2 - 6 October 2010 through 8 October 2010
ER -