DocumentCode :
1734195
Title :
Modeling, optimization and benchmarking of chip-to-chip electrical interconnects with low loss air-clad dielectrics
Author :
Kumar, Vachan ; Bashirullah, Rizwan ; Naeemi, Azad
Author_Institution :
Sch. of Electr. & Comput. Eng., Georgia Inst. of Technol., Atlanta, GA, USA
fYear :
2011
Firstpage :
2084
Lastpage :
2090
Abstract :
Analytical models are developed for pulse response of chip-to-chip interconnects modeled as lossy transmission lines. The novel compact physical models capture the losses due to skin effect and dielectric losses and predict both the pulse height and width at the receiver. The work presented in this paper demonstrates the use of these models for optimization of data-rate and interconnect geometry in order to minimize energy-per-bit and maximize bandwidth-density simultaneously. Further, the model is applied to explore and quantify the potential benefits of low-k air-clad dielectrics that reduce dielectric losses. The model predicts a maximum improvement of ideal air dielectric over a lossy GETEK dielectric to be roughly 5-6 times in both energy-per bit and bandwidth-density. Additionally, the optimizations provided in the paper set the minimum requirements for optical interconnects in order to outperform electrical interconnects.
Keywords :
electric connectors; integrated circuit interconnections; integrated circuit modelling; integrated circuit testing; optical interconnections; transmission lines; GETEK dielectric; analytical models; chip-to-chip electrical interconnects; dielectric losses; lossy transmission lines; low loss air-clad dielectrics; optical interconnects; skin effect; Bandwidth; Dielectric losses; Optimization; Power transmission lines; Receivers; Skin effect;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electronic Components and Technology Conference (ECTC), 2011 IEEE 61st
Conference_Location :
Lake Buena Vista, FL
ISSN :
0569-5503
Print_ISBN :
978-1-61284-497-8
Electronic_ISBN :
0569-5503
Type :
conf
DOI :
10.1109/ECTC.2011.5898805
Filename :
5898805
Link To Document :
بازگشت