A fixed-point implementation of matrix inversion using Cholesky decomposition

被引:0
作者
Burian, A [1 ]
Takala, J [1 ]
Ylinen, M [1 ]
机构
[1] Tampere Univ Technol, Inst Digital & Comp Syst, FIN-33101 Tampere, Finland
来源
PROCEEDINGS OF THE 46TH IEEE INTERNATIONAL MIDWEST SYMPOSIUM ON CIRCUITS & SYSTEMS, VOLS 1-3 | 2003年
关键词
D O I
暂无
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Fixed-point simulation results are used for the performance measure or inverting matrices using the Cholesky decomposition. The working space is reduced significantly by grouping signals using hardware sharing and a variable position of the radix point. The hardware cost of the functional units Is reduced by 25% when compared to floating-point.
引用
收藏
页码:1431 / 1434
页数:4
相关论文
共 11 条
[1]  
BARNES C, 1985, IEEE T ACOUSTIC SIGN, V33, P332
[3]   Truncation noise in fixed-point SFGs [J].
Constantinides, GA ;
Cheung, PYK ;
Luk, W .
ELECTRONICS LETTERS, 1999, 35 (23) :2012-2014
[4]  
HARTLEY R, 1995, DIGIT SERIAL COMPUTA
[5]  
Higham N. J., 1996, ACCURACY STABILITY N
[6]   SYNTHESIS OF MINIMUM ROUNDOFF NOISE FIXED-POINT DIGITAL-FILTERS [J].
MULLIS, CT ;
ROBERTS, RA .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS, 1976, 23 (09) :551-562
[7]   GENERALIZED LINEAR MODELS [J].
NELDER, JA ;
WEDDERBURN, RW .
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES A-GENERAL, 1972, 135 (03) :370-+
[8]   NECESSARY AND SUFFICIENT CONDITION FOR QUANTIZATION ERRORS TO BE UNIFORM AND WHITE [J].
SRIPAD, AB ;
SNYDER, DL .
IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1977, 25 (05) :442-448
[9]  
Stewart GW, 1996, Afternotes on Numerical Analysis, V49, DOI 10.1137/1.9781611971491
[10]   Simulation-based word-length optimization method for fixed-point digital signal processing systems [J].
Sung, WY ;
Kum, KI .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1995, 43 (12) :3087-3090