60GHz

1
RF SoS Small Signal Model Extraction
Mr. Zongru (Jerry) Liu Dr. Efstratios
(Stan) Skafidas Prof. Rob Evans National ICT
Australia Limited Dept of Electrical and
Electronic Engineering University of Melbourne
2
Talk Outline

• Why is Small Signal Model Important
• Overview of the Procedure
• Calibration Method
• 2-step De-embedding
• Extrinsic Parameter Extraction
• Intrinsic Model Extraction
• Results and Conclusion

3
Why is Small Signal Model Important

• 7GHz spectrum being made available worldwide
• Targeting low-cost, high-data-rate communication
  applications using CMOS technology
• Require accurate active and passive models at
  these frequencies
• Foundry supplied process development kit models
  do not support these frequencies.
• Usual technique is prone to failure especially at
  these higher frequencies

4
Overview of the Procedure
Silicon on Saphire transistor
5
Overview of the Procedure
Transistor with parasitic components
illustrated
6
Overview of the Procedure
Transistor equivalent circuit
7
Overview of the Procedure
8
Calibration Method

• Lab setup
• using a calibration substrate and a self
  consistent calibration model SUSS LRM

9
Overview of the Procedure
10
2-step De-embedding
Pads parasitic have to be removed
11
2-step De-embedding
Open and short structures build on chip to
extract PADs parasitic
12
2-step De-embedding

• The open standard can be used to determine the
  parallel parasitics Yp1, Yp2, Yp3
• The short standard with the shunt parasitics Yp1,
  Yp2, Yp3 and series parasitics Z1, Z2, and Z3
  surrounding the transistor can be used to
  determine and remove the pad parasitics

13
Overview of the Procedure
14
Extrinsic Parameter Extraction
The inclusion of the short-b permits an
accurate determination of the extrinsic elements
15
Extrinsic Parameter Extraction

• The lumped extrinsic elements are extracted by
  performing a constrained least square fitting
• The constraint is to ensure that the values of
  inductance and resistance are positive

16
Overview of the Procedure
17
Intrinsic Parameter Extraction

• extrinsic elements must be subtracted to get the
  intrinsic Y-matrix (Yin)
• use right hand side formulas to get the initial
  intrinsic elements
• perform least square fittings to get the
  intrinsic parameters

18
Intrinsic Parameter Extraction

• Equations are derived from the small signal
  intrinsic model where is the estimates of
  the intrinsic Y-matrix.
• The cost function is minimized subject to the
  constraint that all the extracted components are
  positive

19
Talk Outline

• Why is Small Signal Model Important
• Overview of the Procedure
• Calibration Method
• 2-step De-embedding
• Extrinsic Parameter Extraction
• Intrinsic Model Extraction
• Results and Conclusion

20
Results
Extrinsic parameters
Intrinsic parameters
21
Results

• the measured and de-embedded S-parameters of one
  of our transistors
• The transistor is 3 fingers with finger width
  8um, channel length 0.25 um. The bias condition
  is Vds 1.5V, Vgs 0.8V.

22
Results

• the simulated S-parameter using Cadence Spectre
  versus the measurements after de-embedding
• Good agreement is obvious from the content

23
Conclusion

• A new extraction scheme has been demonstrated
  which, after introduction of new structure
  short-b, allows accurate and effective extraction
  of all small-signal circuit elements values from
  S-parameters measurements up to 65 GHz
• The least square optimization technique is
  applied to compensate for the noisiness of the
  measurements
• A two step de-embedding method has also been
  demonstrated
• The simulation results also show very well fit
  with the measurements

24
References

• R. Broderson, Cmos for Ultra Wideband and 60 GHz
  Communication, ISSCC Paper 24.4, pp. 440-441,
  Feb 2004
• Berny, A.D., Niknejad, A.M., Meyer, R.G , A
  1.8-GHz LC VCO with 1.3-GHz tuning range and
  digital amplitude calibration IEEE J.
  Solid-State Circuits, vol. 40,  No 4,  pp.909
  917 April 2005
• R. Anholt and S.Swirhun, Equivalent-circuit
  parameter extraction for cold GaAs MESFETs,
  IEEE Tran. Microwave Theory Tech, vol. 39, pp.
  1243-1247, 1991
• A. Bracale, V. Ferlet-Cavrois, N. Fel, D.
  Pasquet, J.L. Gautier, J.L. Pelloie and J. Du
  Port de Poncharra, A New Approach for SOI
  Devices Small-Signal Parameters Extraction,
  Analog Integrated Circuits and Signal Processing,
  25, 157-169, 2000
• J.P. Raskin, R. Gillon, J. chen, D.V. Janvier and
  J.P. Colinge, Accurate SOI MOSFET
  Characterization at Microwave Frequencies for
  Device Performance Optimization and Analog
  Modeling, IEEE Trans. on Electron Devices, vol.
  45, No. 5, May 1988
• M.C.A.M. Koolen, J.A.M. Geelen and M.P.J.G.
  Versleijen, An Improved De-embedding Technique
  for On-wafer High-Frequency Characterization,
  IEEE 1991 Bipolar Circuit and Technology Meeting
  8.1.
• G. Dambrine, A. Cappy, F. Heliodore, and E.
  Playez, A new method for determining the FET
  small-signal equivalent circuit, IEEE Trans.
  Microwave Theory Tech, vol. 36 , No. 7,
  pp.1151-1159, July 1988.