Work place: Department of Electrical and Computer Engineering, Graduate University of Advanced Technology, Kerman, Iran
E-mail: m.faghih@kgut.ac.ir
Website:
Research Interests: Planning and Scheduling, Image Processing, Data Structures and Algorithms, Analysis of Algorithms
Biography
Mohammad Mehdi Faghih received The Ph.D. degree from the Department of Electrical and Computer Engineering at Shahid Beheshti University (SBU), Tehran, Iran, in 2013. Currently, he is an assistant professor at Graduate University of Advanced Technology (KGUT). His research interests include scheduling algorithms, software security and (color) image processing.
By Mohammad Ali Shafieabadi Fazel Sharifi Mohammad Mehdi Faghih
DOI: https://doi.org/10.5815/ijmecs.2019.08.04, Pub. Date: 8 Aug. 2019
In many classes of applications, mainly in signal and image processing applications, decreasing the static power of computational circuits is a challenge. Multipliers are typically located on the critical path of such systems. A promising approach for energy-efficient design of digital systems is approximate or inexact computing. On the other hand, leakage power and limited scalability become serious obstacles that prevent the continuous miniaturization of the conventional CMOS-based logic circuits. Spin-based devices are considered as promising alternatives for CMOS technology due to their proper characteristics such as near-zero current leakage, sustainability, integrity, low standby power. In this paper a new low power approximate 4-2 compressor is presented which is implemented using spintronic devices. The proposed design is utilized in a multiplier tree for image processing applications. We have simulated and compared the proposed design with state-of-the-art designs in both quantitative and qualitative metrics. The simulation results show that the proposed design has 92% and 188% lower power consumption and PDP, respectively compared to the best state-of-the-art design.
[...] Read more.Subscribe to receive issue release notifications and newsletters from MECS Press journals