CORDIC and SVD implementation in digital hardware

Author

Szecówka, Przemyslaw M. ; Malinowski, Piotr

Author_Institution

Fac. of Microsyst. Electron. & Photonics, Wroclaw Univ. of Technol., Wrocław, Poland

fYear

2010

Firstpage

237

Lastpage

242

Abstract

Singular Value Decomposition is classified among the most effective numeric methods of matrices inversion. The paper presents a study of hardware implementation of SVD and CORDIC algorithms. Various digital architectures were proposed and compared, including low-cost sequential and high-performance pipelined solutions. Fixed point and floating point arithmetic was considered. The concepts were implemented in VHDL, verified and synthesized with Xilinx tools. Selected approach was physically implemented and tested.

Keywords

fixed point arithmetic; floating point arithmetic; hardware description languages; matrix inversion; singular value decomposition; CORDIC algorithms; SVD implementation; VHDL; Xilinx tools; coordinate rotation digital computer; digital hardware; fixed point arithmetic; floating point arithmetic; matrices inversion; singular value decomposition; Clocks; Delay; Hardware; Matrix decomposition; Radiation detectors; Registers; Singular value decomposition; CORDIC; FPGA; SVD; VHDL; digital; hardware;

fLanguage

English

Publisher

ieee

Conference_Titel

Mixed Design of Integrated Circuits and Systems (MIXDES), 2010 Proceedings of the 17th International Conference

Conference_Location

Warsaw

Print_ISBN

978-1-4244-7011-2

Electronic_ISBN

978-83-928756-4-2

Type

conf

Filename

5551655