Prediction of Power Supply Noise from Switching Activity in an FPGA
Switching current drawn by an integrated circuit (IC) creates dynamic power supply noise on the IC and on the printed circuit board (PCB), which in turn causes jitter in $I$/$O$ signals and reduces the maximum clock frequency. Predicting power supply noise is challenging due to the complexity of determining the dynamic current drawn by the IC and the impedance of the power delivery network. In this paper, a methodology is developed for predicting dynamic power supply noise on the PCB resulting from logic activity in a field-programmable gate array (FPGA). Time-domain switching currents within the FPGA are found by performing power simulations of the implemented logic over small time intervals. A high-frequency model of the die-package-PCB power delivery network is developed based on the inductance and capacitance of the package and die and a cavity model description of the PCB. The technique is shown to accurately predict noise on the PCB in both the time and frequency domains.
L. Ren and T. Li and S. K. Chandra and X. Chen and H. Bishnoi and S. Sun and P. A. Boyle and I. G. Zamek and J. Fan and D. G. Beetner and J. L. Drewniak, "Prediction of Power Supply Noise from Switching Activity in an FPGA," IEEE Transactions on Electromagnetic Compatibility, vol. 56, no. 3, pp. 699-706, Institute of Electrical and Electronics Engineers (IEEE), Jun 2014.
The definitive version is available at http://dx.doi.org/10.1109/TEMC.2013.2293872
Electrical and Computer Engineering
Center for High Performance Computing Research
Keywords and Phrases
Complex networks; Electric impedance; Electric power transmission; Field programmable gate arrays (FPGA); Forecasting; Integrated circuits; Models; Organic pollutants; High-frequency modeling; noise; Power delivery network; Power delivery network (PDN); Power integrity; Printed circuit boards (PCB); Switching activities; Time and frequency domains; Printed circuit boards; Impedance; integrated circuit (IC)
International Standard Serial Number (ISSN)
Article - Journal
© 2014 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.