Addition is a key fundamental function for many error-tolerant applications. Approximate addition is considered to be an efficient technique for trading off energy against performance and accuracy. This paper proposes a carry-maskable adder whose accuracy can be configured at runtime. The proposed scheme can dynamically select the length of the carry propagation to satisfy the quality requirements flexibly. Compared with a conventional ripple carry adder and a conventional carry look-ahead adder, the proposed 16-bit adder reduced the power consumption by 54.1% and 57.5%, respectively, and the critical path delay by 72.5% and 54.2%, respectively. In addition, results from an image processing application indicate that the quality of processed images can be controlled by the proposed adder. Good scalability of the proposed adder is demonstrated from the evaluation results using a 32-bit length.
Tongxin YANG
Fukuoka University
Toshinori SATO
Fukuoka University
Tomoaki UKEZONO
Fukuoka University
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.
Copy
Tongxin YANG, Toshinori SATO, Tomoaki UKEZONO, "An Accuracy-Configurable Adder for Low-Power Applications" in IEICE TRANSACTIONS on Electronics,
vol. E103-C, no. 3, pp. 68-76, March 2020, doi: 10.1587/transele.2019LHP0002.
Abstract: Addition is a key fundamental function for many error-tolerant applications. Approximate addition is considered to be an efficient technique for trading off energy against performance and accuracy. This paper proposes a carry-maskable adder whose accuracy can be configured at runtime. The proposed scheme can dynamically select the length of the carry propagation to satisfy the quality requirements flexibly. Compared with a conventional ripple carry adder and a conventional carry look-ahead adder, the proposed 16-bit adder reduced the power consumption by 54.1% and 57.5%, respectively, and the critical path delay by 72.5% and 54.2%, respectively. In addition, results from an image processing application indicate that the quality of processed images can be controlled by the proposed adder. Good scalability of the proposed adder is demonstrated from the evaluation results using a 32-bit length.
URL: https://global.ieice.org/en_transactions/electronics/10.1587/transele.2019LHP0002/_p
Copy
@ARTICLE{e103-c_3_68,
author={Tongxin YANG, Toshinori SATO, Tomoaki UKEZONO, },
journal={IEICE TRANSACTIONS on Electronics},
title={An Accuracy-Configurable Adder for Low-Power Applications},
year={2020},
volume={E103-C},
number={3},
pages={68-76},
abstract={Addition is a key fundamental function for many error-tolerant applications. Approximate addition is considered to be an efficient technique for trading off energy against performance and accuracy. This paper proposes a carry-maskable adder whose accuracy can be configured at runtime. The proposed scheme can dynamically select the length of the carry propagation to satisfy the quality requirements flexibly. Compared with a conventional ripple carry adder and a conventional carry look-ahead adder, the proposed 16-bit adder reduced the power consumption by 54.1% and 57.5%, respectively, and the critical path delay by 72.5% and 54.2%, respectively. In addition, results from an image processing application indicate that the quality of processed images can be controlled by the proposed adder. Good scalability of the proposed adder is demonstrated from the evaluation results using a 32-bit length.},
keywords={},
doi={10.1587/transele.2019LHP0002},
ISSN={1745-1353},
month={March},}
Copy
TY - JOUR
TI - An Accuracy-Configurable Adder for Low-Power Applications
T2 - IEICE TRANSACTIONS on Electronics
SP - 68
EP - 76
AU - Tongxin YANG
AU - Toshinori SATO
AU - Tomoaki UKEZONO
PY - 2020
DO - 10.1587/transele.2019LHP0002
JO - IEICE TRANSACTIONS on Electronics
SN - 1745-1353
VL - E103-C
IS - 3
JA - IEICE TRANSACTIONS on Electronics
Y1 - March 2020
AB - Addition is a key fundamental function for many error-tolerant applications. Approximate addition is considered to be an efficient technique for trading off energy against performance and accuracy. This paper proposes a carry-maskable adder whose accuracy can be configured at runtime. The proposed scheme can dynamically select the length of the carry propagation to satisfy the quality requirements flexibly. Compared with a conventional ripple carry adder and a conventional carry look-ahead adder, the proposed 16-bit adder reduced the power consumption by 54.1% and 57.5%, respectively, and the critical path delay by 72.5% and 54.2%, respectively. In addition, results from an image processing application indicate that the quality of processed images can be controlled by the proposed adder. Good scalability of the proposed adder is demonstrated from the evaluation results using a 32-bit length.
ER -