DocumentCode :
2668584
Title :
Low leakage multi-Vth technique for sequential circuits at transistor level in nanotechnology
Author :
Rjoub, Abdoul ; Almanasrah, Hassan
Author_Institution :
Comput. Eng. Dept., Jordan Univ. of Sci. & Technol., Irbid, Jordan
fYear :
2010
fDate :
12-15 Dec. 2010
Firstpage :
974
Lastpage :
977
Abstract :
Leakage power is the main dominant source of power dissipation for Sub-100nm VLSI circuits. Various techniques were proposed to reduce the leakage power dissipation; one of these techniques is Multi-Threshold voltage. In this paper, the exact and optimal values of Threshold Voltage (Vth) for each transistor of the design are found for any sequential circuit. This is achieved by applying Artificial Intelligence (AI) Search Algorithm to find the optimum values that get the largest reduction of leakage current. The proposed algorithm exploits the total Slack Time of each transistor´s location and their contribution of leakage current. The proposed algorithm is introduced by AI Heuristic Search, under 22nm BSIM4 foundries predictive model. The proposed approach saves around 80% of the Sub-threshold Leakage current without degrading the performance of the circuit.
Keywords :
VLSI; artificial intelligence; integrated circuit design; logic design; nanoelectronics; search problems; sequential circuits; transistors; AI heuristic search algorithm; BSIM4 foundries predictive model; VLSI circuits; artificial intelligence search algorithm; leakage power dissipation; multithreshold voltage; nanotechnology; sequential circuits; transistor level; Artificial intelligence; Benchmark testing; MOSFET circuits; System-on-a-chip; Artificial Intelegence; Leakage Current; Low Power; Multi-Threshold Technique; Nanotechnology; SPICE Parameters;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electronics, Circuits, and Systems (ICECS), 2010 17th IEEE International Conference on
Conference_Location :
Athens
Print_ISBN :
978-1-4244-8155-2
Type :
conf
DOI :
10.1109/ICECS.2010.5724676
Filename :
5724676
Link To Document :
بازگشت