Efficient CORDIC-Based Processing Elements in Scalable Complex Matrix Inversion
Summary :
In this paper we apply angle recoding to the CORDIC-based processing elements in a scalable architecture for complex matrix inversion. We extend the processing elements from the scalable real matrix inversion architecture to the complex domain and obtain the novel scalable complex matrix inversion architecture, which can significantly reduce computational complexity. We rearrange the CORDIC elements to make one half of the processing elements simple and compact. For the other half of the processing elements, the efficient use of angler recoding reduces the number of microrotation steps of the CORDIC elements to 3/4. Consequently, only 3 CORDIC elements are required for the processing elements with full utilization.
- Publication
- IEICE TRANSACTIONS on Fundamentals Vol.E97-A No.5 pp.1144-1148
- Publication Date
- 2014/05/01
- Publicized
- Online ISSN
- 1745-1337
- DOI
- 10.1587/transfun.E97.A.1144
- Type of Manuscript
- LETTER
- Category
- Algorithms and Data Structures
Authors
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Huan HE, Feng YU, Bei ZHAO, "Efficient CORDIC-Based Processing Elements in Scalable Complex Matrix Inversion" in IEICE TRANSACTIONS on Fundamentals,
vol. E97-A, no. 5, pp. 1144-1148, May 2014, doi: 10.1587/transfun.E97.A.1144.
Abstract: In this paper we apply angle recoding to the CORDIC-based processing elements in a scalable architecture for complex matrix inversion. We extend the processing elements from the scalable real matrix inversion architecture to the complex domain and obtain the novel scalable complex matrix inversion architecture, which can significantly reduce computational complexity. We rearrange the CORDIC elements to make one half of the processing elements simple and compact. For the other half of the processing elements, the efficient use of angler recoding reduces the number of microrotation steps of the CORDIC elements to 3/4. Consequently, only 3 CORDIC elements are required for the processing elements with full utilization.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E97.A.1144/_p
Copy
@ARTICLE{e97-a_5_1144,
author={Huan HE, Feng YU, Bei ZHAO, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Efficient CORDIC-Based Processing Elements in Scalable Complex Matrix Inversion},
year={2014},
volume={E97-A},
number={5},
pages={1144-1148},
abstract={In this paper we apply angle recoding to the CORDIC-based processing elements in a scalable architecture for complex matrix inversion. We extend the processing elements from the scalable real matrix inversion architecture to the complex domain and obtain the novel scalable complex matrix inversion architecture, which can significantly reduce computational complexity. We rearrange the CORDIC elements to make one half of the processing elements simple and compact. For the other half of the processing elements, the efficient use of angler recoding reduces the number of microrotation steps of the CORDIC elements to 3/4. Consequently, only 3 CORDIC elements are required for the processing elements with full utilization.},
keywords={},
doi={10.1587/transfun.E97.A.1144},
ISSN={1745-1337},
month={May},}
Copy
TY - JOUR
TI - Efficient CORDIC-Based Processing Elements in Scalable Complex Matrix Inversion
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1144
EP - 1148
AU - Huan HE
AU - Feng YU
AU - Bei ZHAO
PY - 2014
DO - 10.1587/transfun.E97.A.1144
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E97-A
IS - 5
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - May 2014
AB - In this paper we apply angle recoding to the CORDIC-based processing elements in a scalable architecture for complex matrix inversion. We extend the processing elements from the scalable real matrix inversion architecture to the complex domain and obtain the novel scalable complex matrix inversion architecture, which can significantly reduce computational complexity. We rearrange the CORDIC elements to make one half of the processing elements simple and compact. For the other half of the processing elements, the efficient use of angler recoding reduces the number of microrotation steps of the CORDIC elements to 3/4. Consequently, only 3 CORDIC elements are required for the processing elements with full utilization.
ER -
English
