Research Article
Approximate Computing Based Low Power Image Processing Architecture for Intelligent Satellites
@INPROCEEDINGS{10.1007/978-3-030-69072-4_28, author={Zhixi Yang and Rong Lv and Xianbin Li and Jian Wang and Jun Yang}, title={Approximate Computing Based Low Power Image Processing Architecture for Intelligent Satellites}, proceedings={Wireless and Satellite Systems. 11th EAI International Conference, WiSATS 2020, Nanjing, China, September 17-18, 2020, Proceedings, Part II}, proceedings_a={WISATS PART 2}, year={2021}, month={2}, keywords={Approximate multiplier Approximate adder Low power Real time Image processing Intelligent satellite}, doi={10.1007/978-3-030-69072-4_28} }- Zhixi Yang
Rong Lv
Xianbin Li
Jian Wang
Jun Yang
Year: 2021
Approximate Computing Based Low Power Image Processing Architecture for Intelligent Satellites
WISATS PART 2
Springer
DOI: 10.1007/978-3-030-69072-4_28
Abstract
Approximate computing is an innovative circuit paradigm for lower power and real time image processing architecture within an intelligent satellite. Multiplication and addition are often fundamental functions for many image processing applications. Based on previous approximate compressor designs, a recursive type multiplier is first proposed. A reduced gate-level complexity full adder is then proposed. Extensive simulation results show that the proposed designs achieve significant reductions in area, power and delay compared with exact recursive multiplier and adders, as well as other approximate designs found in the technical literature. An image processing application is performed to further show that the performance of the proposed approximate designs for image processing achieves a very good accuracy (measured by the peak signal to noise ratio) as well as substantial reductions in power dissipation and delay.
