Title :
Reconfigurable particle filter design using dataflow structure translation
Author :
Sangjin Hong; Xiaoyao Liang;P.M. Djuric
Author_Institution :
Dept. of Electr. & Comput. Eng., State Univ. of New York, Stony Brook, NY, USA
fDate :
6/26/1905 12:00:00 AM
Abstract :
This paper presents reconfigurable particle filter design, which provides a capability of selecting a single particle filter from multiple particle filter realizations. The execution of the design is based on block level pipelining where data transfer between processing blocks is effectively controlled by autonomous controllers. With a simple switching mechanism that allows transformation of dataflow structure in addition to autonomous buffer controller, any desired particle filter can be performed. Two target particle filters, based on SIRF and GPF, are realized. From the execution characteristics obtained from the FPGA implementation, overall controller structure is derived according to the methodology and verified using Verilog and SystemC.
Keywords :
"Particle filters","Particle tracking","Pipeline processing","Field programmable gate arrays","Hardware design languages","Filtering","Design methodology","Gaussian noise","Control systems","Signal processing"
Conference_Titel :
Signal Processing Systems, 2004. SIPS 2004. IEEE Workshop on
Print_ISBN :
0-7803-8504-7
DOI :
10.1109/SIPS.2004.1363071