GSFAP Adaptive Filtering Using Log Arithmetic for Resource-Constrained Embedded Systems

0341115 - ÚTIA 2011 RIV US eng J - Článek v odborném periodiku
Tichý, Milan - Schier, Jan - Gregg, D.
GSFAP Adaptive Filtering Using Log Arithmetic for Resource-Constrained Embedded Systems.
ACM Transactions on Embedded Computing Systems. Roč. 9, č. 3 (2010), s. 1-31. ISSN 1539-9087. E-ISSN 1558-3465
Grant CEP: GA MŠMT 7H09005
Výzkumný záměr: CEZ:AV0Z10750506
Klíčová slova: FPGA * DSP * logarithmic arithmetic * affine projection
Kód oboru RIV: BD - Teorie informace
Impakt faktor: 1.057, rok: 2010
http://library.utia.cas.cz/separaty/2010/ZS/tichy-0341115.pdf

This article shows that FPGA devices are a highly suitable platform for more computationally intensive adaptive algorithms. We present an optimized core which implements GSFAP. GSFAP is an algorithm with far superior adaptation properties than NLMS, and with only slightly higher computational complexity. To further optimize resource requirements we use logarithmic arithmetic, rather than conventional floating point, within the custom core.
Trvalý link: http://hdl.handle.net/11104/0184199