DocumentCode :
2736246
Title :
An analytical delay model for SRAM-based FPGA interconnections
Author :
Feng, Zhou ; Zhijun, Huang ; Jiarong, Tong ; Pushan, Tang
Author_Institution :
CAD Inst., Fudan Univ., Shanghai, China
fYear :
1999
fDate :
18-21 Jan 1999
Firstpage :
101
Abstract :
In an SRAM-based FPGA, MOS transistors connect wire segments to construct interconnections between CLBs, resulting in large and unpredictable path delays. So it is necessary to be able to estimate interconnection delays quickly and accurately in order that performance-driven layout and analysis algorithms can achieve high quality. Because the effective channel resistance of a MOS transistor changes with the voltage on a transistor´s source pole, general methods for wire nets´ path delay estimation, in which wire resistance is always a fixed value, will never be applicable, while SPICE would be too computationally expensive to be used in layout optimization. In this paper, an analytical delay model of FPGA interconnection under ramp input is put forward, and closed-form formulas for delay estimation are proposed. Compared with SPICE, our algorithm has been proved to be fast and accurate enough
Keywords :
circuit analysis computing; delay estimation; field programmable gate arrays; integrated circuit interconnections; integrated circuit layout; integrated circuit modelling; logic CAD; SRAM-based FPGA interconnections; analytical delay model; closed-form formulas; interconnection delay estimation; layout optimization; path delay estimation; performance-driven analysis algorithms; performance-driven layout algorithms; ramp input; Algorithm design and analysis; Analytical models; Delay estimation; Field programmable gate arrays; MOSFETs; Optimization methods; Performance analysis; SPICE; Voltage; Wire;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Design Automation Conference, 1999. Proceedings of the ASP-DAC '99. Asia and South Pacific
Conference_Location :
Wanchai
Print_ISBN :
0-7803-5012-X
Type :
conf
DOI :
10.1109/ASPDAC.1999.759723
Filename :
759723
Link To Document :
بازگشت