generating spectrally rich data sets using adaptive band ... · data sets using adaptive band...
TRANSCRIPT
![Page 1: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/1.jpg)
Generating Spectrally Rich Data Sets Using Adaptive
Band Synthesis Interpolation
James C. RautioSonnet Software, Inc.
WFA: Microwave Component DesignUsing Optimization Techniques
June 2003
![Page 2: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/2.jpg)
ABS Overview
Interpolation PrimerInterpolation Primer
Data Points
Interpolated Points
y = a + bx
Linear interpolation– Draw a straight
line between each data point.
– Interpolated points fall on that line.
![Page 3: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/3.jpg)
ABS Overview
Interpolation PrimerInterpolation Primer
Cubic spline– Take four data
points.– Calculate cubic
curve that goes exactly through all four points.
– Interpolated data lies on that line.
Data Points
Interpolated Points
y = a + bx + cx2 + dx3
![Page 4: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/4.jpg)
ABS Overview
Interpolation PrimerInterpolation Primer
...1...
221
2210
++++++
=xbxbxaxaay
Cubic spline uses y = a + bx + cx2 + dx3.Taking a hint from Laplace transform theory, try:
This is known as a Padé polynomial.
![Page 5: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/5.jpg)
ABS Overview
ABS ABS AlgorithimAlgorithim
1. Analyze first, last, and mid frequency.2. Form several interpolation models.3. Estimate interpolation error.4. Find frequency of worst error.5. If worst error is small enough, quit.6. Otherwise, analyze worst frequency
and return to step 2.
![Page 6: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/6.jpg)
ABS Overview
PadPadéé Polynomial ProblemsPolynomial Problems
...1...
221
2210
++++++
=xbxbxaxaay
Perfect for lumped circuits, but band limited for distributed circuits.Matrix has terms like (freq)N, matrix precision problem likely for large N.Must be able to estimate interpolation error so we know where to take next data point and when to quit.
![Page 7: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/7.jpg)
ABS Overview
Bandwidth SolutionsBandwidth Solutions
Extract additional internal information from moment matrix.Estimate zero locations.Build much higher order Padépolynomial model.Much wider bandwidth now possible.
![Page 8: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/8.jpg)
ABS Overview
Interpolation ErrorInterpolation Error
Interpolation error is difference between the interpolation and the correct answer.To estimate error, do two different interpolations on same data.Difference between interpolations is error estimate.
![Page 9: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/9.jpg)
ABS Overview
Interpolation ErrorInterpolation Error
With 4 data points:– Do interpolation with 4 data points.– Do interpolation with 3 data points.– Difference is error estimate.Next analysis frequency goes at frequency with largest error estimate.When largest error estimate is very small, all done!
![Page 10: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/10.jpg)
ABS Overview
Interpolation ErrorInterpolation Error
True error can sometimes be up to 20 dB worse than estimated error.For a good plot, true error must be 20 dB less than the data being plotted.Thus, we continue until the error estimate is over 40 dB less than the data being plotted.
![Page 11: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/11.jpg)
ABS Overview
Interpolation Error ExampleInterpolation Error Example
At 0 dB (mag=1.0), error is 40 dB down if mag is ±0.01.
1.0±0.01
Mag
nitu
de (d
B)
0
-20
-40
-80
-60
Frequency
![Page 12: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/12.jpg)
ABS Overview
Interpolation Error ExampleInterpolation Error Example
At -40 dB (mag=0.01), error is 40 dB down if mag is ±0.0001.
0.01±0.0001
Mag
nitu
de (d
B)
0
-20
-40
-80
-60
Frequency
![Page 13: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/13.jpg)
ABS Overview
Interpolation Error ExampleInterpolation Error Example
At -80 dB (mag=0.0001), error is 40 dB down if mag is ±0.000001. 0.0001±
0.000001M
agni
tude
(dB)
0
-20
-40
-80
-60
Frequency
![Page 14: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/14.jpg)
ABS Overview
Interpolation Error ExampleInterpolation Error Example
We ran 150 test circuits and plotted true versus interpolated.All plots visually identical when estimated error 40 dB or more below data.
Est. error upper limit
Mag
nitu
de (d
B)
0
-20
-40
-80
-60
Frequency
![Page 15: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/15.jpg)
ABS Overview
Hairpin FilterHairpin Filter
300 data points.
Sonnet ABS interpolated from analysis at four frequencies.Results visually identical.
![Page 16: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/16.jpg)
ABS Overview
Hairpin FilterHairpin Filter
Phase also gives visually identical results.
![Page 17: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/17.jpg)
ABS Overview
DiplexerDiplexer7-Layer LTCC Design
Antenna In
Band 1 out
Band 2 out
![Page 18: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/18.jpg)
ABS Overview
DiplexerDiplexer7-Layer LTCC Design
9 Discrete EM Frequencies required, 320 frequencies provided over 8x Bandwidth
![Page 19: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/19.jpg)
ABS Overview
Band Pass FilterBand Pass FilterHigh Temp Superconductor Microstrip Filter
HTSC filter inside rectangular housing.Precise enclosure effects are very important and are included. (Microwaves &RF, Dec 98, pp. 119-130)
ABS yields detailed response.
![Page 20: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/20.jpg)
ABS Overview
Band Pass FilterBand Pass FilterHigh Temp Superconductor Microstrip Filter
Four frequency ABS analysis.Total time: 6 minutes on 1 GHz laptop PCMeasured versus calculated.Filter courtesy George Matthaei.
![Page 21: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/21.jpg)
ABS Overview
Big Low Pass FilterBig Low Pass FilterExtremely fine subsectioning used.2m 3s per frequency (1.5 GHz P4).Note fine geometry along center line.
![Page 22: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/22.jpg)
ABS Overview
Big Low Pass FilterBig Low Pass Filter
Sonnet ABSLDR-EM
Sonnet ABS uses 13 frequencies, 30 minutes.Lower dynamic range analysis (LDR-EM) needs 117 frequencies, about 2.5 days.EM dynamic range is important!
![Page 23: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/23.jpg)
ABS Overview
Spiral SplitterSpiral Splitter
Sonnet ABS
Non-ABS
Sonnet ABS uses 8 frequencies.Non-ABS approach uses 23.Information from MoM matrix is important!
![Page 24: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/24.jpg)
ABS Overview
Spiral Inductor on SiliconSpiral Inductor on SiliconEight turn spiral.2-layer thick metal model.Conformal meshing.Measured and calculated essentially identical.ABS w/6 frequencies.
S11 Meas& Calc
S21 Calc
S21 Meas
Data courtesy of Motorola, circuit uses Motorola High Voltage IC (HVIC) Si RF-LDMOS process.
![Page 25: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/25.jpg)
ABS Overview
Optimization ExampleOptimization ExampleOptimize low freq (0.5 GHz) inductance for S21 phase from –19.8° to –25°.Vary parameter “Stretch” from 50 to 100 microns.
![Page 26: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/26.jpg)
ABS Overview
Optimization ResultOptimization Result
Start S21 Ang
Finish S21 Ang
Finish S21 Mag
Start S21 Mag
Requires 7 EM analyses, 26 minutes per analysis, 1.3 GHz P4.1.6 µm cell size, 6.4 µm line width.Optimal “Stretch” = 57.4 µm.
![Page 27: Generating Spectrally Rich Data Sets Using Adaptive Band ... · Data Sets Using Adaptive Band Synthesis Interpolation James C. Rautio ... Antenna In Band 1 out Band 2 out. ... 2m](https://reader030.vdocuments.us/reader030/viewer/2022032615/5d27be5888c993b7508d1479/html5/thumbnails/27.jpg)
ABS Overview
ConclusionConclusion
Sonnet ABS uses Padé polynomial.Bandwidth problems overcome by making extensive use of internal EM data.High dynamic range EM software is critical!Octave bandwidth typical with < 6 analyses.Decade bandwidth typical with < 15 analysis.Fast, spectrally rich broad band data sets allow detailed optimization goals and result presentation.