Chapter 3 Design and Measurement of Components
3.1 Design and Measurement of the I/Q Hybrid Mixer
3.2.1 Simulation and Measurement of the DGS
; W=1.15mm A. One-section DGS
S11 S21
a=b=3.5mm ; g=0.2mm
RO4003 εr=3.38 ; h=0.508mm
h
a b
a
b g
w w
(a)
; (b) result (b)
Figure 3.10 EM-simulation of the one-section DGS (a) 3-D view
`
S21 S11
(a) Top side
Figure 3.11 Measurement result of the one-section DGS (b) Bottom side
Figure 3.12 Photograph of the one-section DGS
B. Two-section DGS
a=b=3mm ; g=0.2mm ; W=1.15mm d=6mm ; h=0.508mm
RO4003 εr=3.38
(a) (b)
Figure 3 ; (b) result
3.14 Measurement result of the two-section DGS
Figure 3.1
S11 S21
h a
b a
b
w
g d
W
.13 EM-simulation of the two-section DGS (a) 3-D view
(a) Top side
Figure
S11 S21
(b) Bottom side 5 Photograph of the two-section DGS
C. Three-section DGS
a=b=3.5mm ; g=0.2mm ; W=1.15mm
d=6mm ; h=0.508mm
RO4003 εr=3.38 S11
S21
h a b
b a
w
g d
W
(a) (b)
Figure iew; (b) result
op side
three-section DGS
3.16EM-simulation of the three-section DGS (a) 3-D v
(a) T
Figure 3.17 Measurement result of the three-section DGS
S21 S11
(b) Bottom side
Figure 3.18 Photograph of the
The method of modeling the DGS had been presented in section 2.3. Following up this method, the equivalent circuit of the two-section DGS can be modeled as the
π -type symmetrical two-port circuit as shown in Fig.3.19.
Fig.3.20 shows the simulation result for the equivalen
t circuit of the two-section
DGS that is very similar to EM-simulation result as shown in Fig. 3.13(b).
Figure 3.19 The equivalent circuit of the two-section DGS
CAP C=
ID=
0.2081 pF Cg IND L=
ID=
1.695 nH Lg
CAP C=
ID=
0.7884 pF Cp
CAP C=
ID=
0.7884 pF Cp RES R=
ID=
8218 Ohm Rg
RES R=
ID=
2533 Ohm RES Rp
R=
ID=
2533 Ohm Rp
PORT P= 2 P= 1 PORT
Figure 3.20 Simulation result for the equivalent circuit of the two-section DGS
3.2.2 r
3.23 Measurement result of 5.25GHz
An unequal power divider can be designed using the factor -output power ratio [16].
In this thesis, the output power ratio of ed to
be 3:1. The simulation and measurement results of the S-parameters , of show as 1.78dB, 6.25dB and 2.12dB, 7.01dB as shown in Fig.3.21 and Fig3.23 respectively.
Simulation and Measurement of 5.25GHz Unequal Power Divide
Figure 3.22 Photograph of 5.25GHz unequal power divider
Port 1 Port 3 Port 2
Figure
unequal power divider
k2
5.25GHz unequal power divider is assum S21 S31
3.2.3
Simulation and Measurement of 5.25GHz Low Pass Filter
PCB :
RO4003 εr=3.38 with thickne
a) (b)
) result
Figure 3.26 Measurement result of 5.25GHz LPF
5.25GHz Low Pass Filter (LPF) using microstrip line circuitry was used in order to obt
0.508 mm ss
(
Figure 3.24 EM-simulation of 5.25GHz LPF (a) 3-D view; (b
Figure 3.25 Photograph of 5.25GHz LPF
S11 S21
S21 S11
ain a pure signal at the RF input of ADF4106. The results of EM-simulation and measurement for 5.25GHz LPF are shown in Fig.3.24 and Fig.3.26 respectively.
3.2.4 Simulation and Measurement of 5.25GHz VCO with and without DGS
Figure 3.27 Schematic of the negative-impedance using the BFG425W A. Simulation of the negative-impedance circuit
CAP C=
ID=
0.8 pF
C1 LOAD
Z=
ID=
50 Ohm Z1
MLSC
L=
W=
ID=
10.4 mm 2.5 mm TL1 CAP MLIN
C=
ID=
0.8 pF C2
L=
W=
ID=
1.5 mm 1.27 mm
TL2 C
B
E 1
2
3 SUBCKT ID=BFG425W PORT
Z=
P=
50 Ohm 1
short stub
Figure 3.28 Simulation result of the negative-impedance circui
The bipolar transistor BFG425W (Philips) is used to act as the active device on the cir
t
cuit of 5.25GHz VCO.A short stub was connected to the emitter of the BFG425W in order to obtain a negative real part of input impedance as shown in Fig.3.27. Simulation result of the negative-impedance circuit is shown in Fig.3.28.
B. Simulations of the resonator with and without DGS
Figure 3.29 Schematic of the resonator (a) with DGS (b) without DGS
ID=DGS_S2P PORT
Z=
( a ) real part of input impedance ( b ) imaginary part of input impedance Figure 3.30 Simulation results of the
TOSHIBA show
resonator with and without DGS
1SV285 varactor diode is used on the resonator of 5.25GHz VCO as n in Fig.3.29 In the circuit of the resonator with DGS, the S2P file of the two-section DGS is extracted from EM-simulation. Fig.3.30 shows the simulation results of the resonator with and without DGS that the imaginary part of the resonator with DGS has higher slope than the imaginary part of the resonator without DGS. It implies that the resonator with DGS has a higher quality factor.
C. Measurement result of 5.25GHz VCO with DGS
Table 3.3 Measurement result of 5.25GHz VCO with DGS
Figure 3.31 The curve of Frequency vs. Tuning Voltage for 5.25GHz VCO with DGS
The performance of 5.25GHz VCO with DGS has been measured as shown in Table3.3. Fig.3.31 shows the curve of Frequency vs. Tuning Voltage for 5.25GHz VCO with DGS.
Vt (V) f0 (MHz) outptu power(dBm)
0 5181 5.5
0.5 5204 5.5
1 5225 5.5
1.5 5248 5.67
2 5272 5.67
2.5 5297 5.67
3 5326 5.83
3.5 5357 6
4 5390 6.33
4.5 5423 6.5
5 5449 6.67
Sensitivity(MHz/V) 44
47 54
64 59
5150 5200 5250 5300 5350 5400 5450 5500
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Vt(V)
f0(MHz)
D. Measurement result of 5.25GHz VCO without DGS
Figure 3.32The curve of Frequency vs. Tuning Voltage for 5.25GHz VCO without DGS
The performance of 5.25GHz VCO without DGS has been measured as shown in Table3.4. Fig.3.32 shows the curve of Frequency vs. Tuning Voltage for 5.25GHz VCO without DGS.
Table 3.4 Measurement result of 5.25GHz VCO without DGS Vt (V) f0 (MHz) outptu power(dBm)
0 5189 5
0.5 5205 5.33
1 5220 5.33
1.5 5235 5.33
2 5250 5.5
2.5 5266 5.5
3 5283 5.5
3.5 5302 5.5
4 5322 5.5
4.5 5344 5.67
5 5362 5.67
Sensitivity(MHz/V) 31
30 33
39 40
5150 5200 5250 5300 5350 5400
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Vt(V)
f0(MHz)
3.2.5 Measurement of 5.25GHz PLL Frequency Synthesizer with and without DGS Measurement results of 5.25GHz PLL Frequency Synthesizer with DGS
( a ) f0 = 5200MHz ( b ) f0 = 5250MHz
( c ) f0 = 5300MHz ( d ) f0 = 5350MHz Figure 3.33 Measurement results of the phase noise with DGS A.
( a)
Figure 3.34 Measurement results with DGS ( a )tuning range; ( b ) spectrum
( a) Top side ( b ) Bottom side Figure 3.35 Photograph of 5.25GHz PLL frequency synthesizer with DGS
( b )
B. Measurement results of 5.25GHz PLL Frequency Synthesizer without DGS
(
( a ) f0 = 5200MHz ( b ) f0 = 5250MHz
c ) f0 = 5300MHz ( d ) f0 = 5350MHz Figure 3.36 Measurement results of the phase noise without DGS
( a ) ( Figure 3.37 Measurement results without
Figu DGS
b )
DGS ( a )tuning range ; ( b ) spectrum
( a) Top side ( b ) Bottom side re 3.38 Photograph of 5.25GHz PLL frequency synthesizer without
Table 3.5 Summary of the measured phase noise with and without DGS
ection 3.2.5 shows measurement results of the phase noise, tuning range and
Fig.3.38 show the photographs of 5.25GHz PL DGS respectively.
able 3.5 shows a summary of the measured phase noise for 5.25GHz PLL frequency synthesizer with and without DGS. It appears that using the two-section DGS can reduced the phase noise by 10-16dB at 100 KHz offset from carrier. Thus, a higher quality factor can be obtained using the two-section DGS.
Frequency (M z)
@10KHz offset (dBc/Hz)
@100KHz offset (dBc/Hz)
@10KHz offset (dBc/Hz)
@100KHz offset (dBc/Hz)
5200 -78.17 -110.17 -67.5 -93.67
5250 -81.67 -109.67 -66.5 -95.5
-92.83 The phase noise
With DGS
The phase noise Without DGS
H
5300 -79.0 -109.5 -67.17
5350 -77.83 -105.17 -66.67 -94.83
S
spectrum for 5.25GHz PLL frequency synthesizer with and without DGS. Fig.3.35 and L frequency synthesizer with and without
T
3.3 Simulation and Measurement of the Frequency Doubler
hort stub that can minimize the required circuit size with a simple bias c
urement result is better than the simulation result about 5.9dB at input power of 4dBm.The photograph of 5.25GHz to 10.5GHz frequency doubler is shown in Fig.3.42. The conversion gain of 5.25GHz to 10.5GHz frequency doubler has been measured for various input power as shown in Table3.6. Fig.3.43 shows the conversion gain and output power vs. input power curve for 5.25GHz to 10.5GHz frequency doubler.
Short stub
FHX35LG
Figure 3.39 Schematic of the frequency doubler using the FHX35LG
The FHX35LG is used on the circuit of 5.25GHz to 10.5GHz frequency doubler to act as the active device as shown in Fig.3.39. The gate voltage is set as zero volts by means of using a s
ondition, namely no requirement for RF choke and negative voltage at the gate terminal. The results of the simulation and measurement for 5.25GHz to 10.5GHz frequency doubler with input power of 4dBm are shown in Fig.3.40 and Fig.3.41 respectively. It appears that the conversion gain of the meas
cy doubler with input power of 4dBm
RFfreq=
m1=-16.654 5.250E9
m1 m2
RFfreq=
m2=3.060 5.250E9 m1 RFfreq=
m1=-16.654 5.250E9 m2
Figure 3.40 Simulation results of the frequen
RFfreq=
m2=3.060 5.250E9
5.210G 5.220G 5.230G 5.240G 5.250G 5.260G 5.270G 5.280G 5.290G
5.200G 5.300G
-15 -10 -5 0
-20 5
RFfreq
Spectrum[1]
m1
Spectrum[2]
Fundamental and Second Harmonic in dBm m2
freq=5.250GHz m1
m1=-16.654
m2
m2=3.060 freq=10.50GHz m1
m1=-16.654
freq=5.250GHz m2
m2=3.060 freq=10.50GHz
2.00G 4.00G 6.00G 8.00G 10.0G 12.0G 14.0G 16.0G 18.0G 20.0G 22.0G 24.0G 26.0G
0.000 28.0G
-40 -30 -20
-50 -10 0 10
freq, Hz
m1
m2 Second Harmonic
Fundamental
( a ) Flatness
Output Spectrum [dBm]
( b ) Spectrum
Spectrum
( a ) Flatness ( b ) Spectrum
Figure 3.41 Measurement results of the frequency doubler with input power of 4dBm
h of 5.25GH 5GHz Frequency Doubler Figure 3.42 Photograp z to 10.
Table 3.6 Measurement results of the frequency doubler with various input power
2
Figure 3.43 The conversion Gain vs. input power curve for the Frequency Doubler
Input power
@5.25GHz
(dBm) Output power(dBm)@10.5GHz* * Conversion Gain(dB)
-10 -5.83 4.17
Conversion Gain(dB) Output power(dBm)@10.5GHz
3
. 10.5GHz CB : O4003
.4 Simulation and Measurement of 10.5GHz Band Pass Filter
A Hairpin Band Pass Filter with N=5 P
R εr=3.38 with
(a) (b)
Figure 3.44 EM-simulation of 10.5GHz Hairpin BPF (N=5) (a) 3-D view; (b) result
Figu
(N=5)
0.508-mm thickness
S11
S21
re 3.45 Photograph of 10.5GHz BPF(N=5)
Figure 3.46 Measurement result of 10.5GHz BPF
26.5mm
S11
S21
B. 10.5GHz Hairpin Band Pass Filter with N=3 PCB :
RO4003 εr=3.38 with 0.508-mm thickness
(a) (b)
F (N=3) (a) 3- ) result
Fi
F(N=3)
Section 3.4 shows the results of simulation and measurement for 10.5GHz hairpin It appears that the rejection of 10.5GHz hairpin
esis, 10.5GHz hairpin BPF with N=3 had been adopted due to its small size.
Figure 3.47 EM-simulation of 10.5GHz Hairpin BP D view; (b
gure 3.48 Photograph of 10.5GHz BPF(N=3) 15mm
S11
S21
S11
S21
Figure 3.49 Measurement result of 10.5GHz BP
BPF with N=3 and N=5 respectively.
BPF with N=5 was deeper than 10.5GHz hairpin BPF with N=3 obviously. But in this th
Chapter 4 Integration and Measurement of the Transceiver
tic of the designed Doppler radar transceiver
ponent had been demonstrated in the chapter 3,
schem
Since the performance of each com
these components are then integrated as a Doppler radar transceiver. Fig.4.1 shows the atic of the designed 10.5GHz Doppler radar transceiver which consists of 5.25GHz PLL frequency synthesizer with DGS, 5.25GHz to 10.5GHz frequency doubler, 10.5GHz Band Pass Filter (BPF), I/Q hybrid mixers, IF amplifiers and voltage regulators.
ADF4106BRU 1 2 3 4 5 6 891011121314 7
15
16 DGNDCECLKDATALE
MUXOUT
DVDD
VPRSET CP CPGND AGND RFIN_B RFIN_A REFIN
AVDD
The designed Doppler radar transceiver has been integrated and fabricated as shown in Fig.4.2. The measurement results of the designed transceiver for the output power of 6dBm and 0dBm are shown in Fig.4.3 and Fig.4.4 respectively. Figure 4.3 (c) shows the output frequency range is from 10.4GHz to 10.8GHz.
Figure 4.2 Photograph of the designed Doppler radar transceiver
(a) Phase noise
Figure 4.3 Measurement results of the designed transceiver with output power of 6 dBm
(b) Spectrum
(c) Flatness
Figure 4.3 ( continued )
(b) spectrum
igure 4.4 Measurement results of the designed transceiver with output power of 0 dBm (a) phase noise
F
Figure 4.5 Photograph of the Doppler radar sensor
The Doppler radar sensor consists of the designed transceiver, an X-Band standard horn antenna, NI DAQCard-6062E with Digital Signal Processing (DSP) and a battery as shown in Fig.4.5 which has been measured for an approaching vehicle at the velocity of 60km ing vehicle at the velocity of 45km/hr respectively. The sample rate is assumed to be 50k samples/sec. Fig.4.6 shows Q channel lead to I channel for an approaching vehicle at the velocity of 60km/hr. Fig.4.7 shows Q channel lag to I channel for a receding vehicle at the velocity of 45km/hr. Fig.4.8 shows the Doppler frequency shift is about 1172Hz and the image rejection is about 29dB for an approaching vehicle
at the ve oppler frequency shift is about -878Hz
X-Band standard horn antenna
The designed transceiver Battery NI DAQCard-6062E
/hr and a reced
locity of 60km/hr. Fig.4.9 shows the D
and the image rejection is about 25dB for a receding vehicle at the velocity of 45km/hr.
Figure 4.6 Measurement result of I/Q channels for an approaching vehicle at the velocity of 60km/hr
velocity of 45km/hr
4 .0 4 .2 4 .4 4 .6 4 .8 5 .0 5 .2 5 .4 5 .6 5 .8 6 .0
- 0 .0 8 - 0 .0 6 - 0 .0 4 - 0 .0 2 0 .0 0 0 .0 2 0 .0 4 0 .0 6 0 .0 8
Amplitude(V)
T im e (m s ) Q ( t ) I(t )
Figure 4.7 Measurement result of I/Q channels for a receding vehicle at the
3.6 3.8 4.0 4.2 4.4 4.6 4.8 5.0 5.2 5.4 5.6 5.8 6.0 6.2 6.4 6.6 0.2
-0.2 -0.1 0.0 0.1
AmlitudeV)
Tim e(m s)
p(
Q(t) I(t)
Image Signal
-10000 -8000 -6000 -4000 -2000 0 2000 4000 6000 8000 10000 -30
-20 -10 0 10 20 30
Amplitude (dB)
Frequency (Hz)
F
a the velocity of 60km/hr
Figur
igure 4.8 Measurement result of the Doppler spectrum (image rejection) for an pproaching vehicle at
Signal Image
-10000 -8000 -6000 -4000 -2000 0 2000 4000 6000 8000 10000 -20
-10 0 10 20 30 40 50
Amplitude(dB)
Frequency(Hz)
e 4.9 Measurement result of the Doppler spectrum (image rejection) for a receding vehicle at the velocity of 45km/hr
Chapter 5 Conclusion and Future Study
An I/Q hybrid mixer is proposed and demonstrated in this thesis which the transmitted signal and the received echo signal pass through the same antenna with no circulator. In addition, the phase noise of oscillator can be reduced by 10-16dB at 100KHz offset from carrier using the DGS. Finally, the designed Doppler radar transceiver has been measured using an X-band standard horn antenna and NI DAQCard-6062E with Digital Signal Processing (DSP) for an approaching vehicle and a receding vehicle. The measurement result shows the validity of distinguishing an approaching vehicle or a receding vehicle and the related velocity.
Due to the 400MHz tunable synthesizing frequency, the designed transceiver
be used with on of multi-path sensors.
transceiver is its big circuit size. It is necessary to igned transceiver with Digital Signal Processing (DSP) can be realized in the MMIC to minimize the circuit size.
can the frequency-scanning antenna for the applicati
The drawback of the designed
reduce the circuit size. In the future, it is worth considering that the des
Refere
[1] B. M
nce
cQueen , J. McQueen , Intelligent Transportation Systems Architecture ,
[2] Zhu Jun; He Peikun; Shi Caicheng; Huang Changzhen;“A Novel Design of Radar Vehicle Detector Based on DSP“, Computational Electromagnetics and Its Boston: Artech House,1999
Applications, 2004. Proceedings.ICCEA 2004. 2004 3rd International Conference [3] Zhang Yanfang , Xia Shouxian ;”The application of PLC on the vehicle detector
System” [J].Journal of Guangdong Automotive and Information Engineering,
[4] Roselli, L.; Alimenti, F.; Comez, M.; Palazzari, V.; Placen 2002,9(4):8-10
tino, F.; Porzi, N.;
Scarponi, A.; “A Cost Driven 24GHz Doppler Radar Sensor Development for Automotive Applications“, 2005 European Microwave Conference , Volume 3, 4-6 Oct. 2005 Page(s):4 pp.
[5] R
Doppler Radar Sensor for Automotive Applications” , Microwave Symposium asshofer, R.H.; Biebl, E.M.;“A Direction Sensitive , Integrated , Low Cost Digest, 1998 IEEE MTT-S International , Volume 2, 7-12 June 1998 Page(s):1055 - 1058 vol.2
[6] David M. Pozar , MICROWAVE ENGINEERING , Third Edition , John Wiley &
Sons , 2005, pp.663-664
] Merrill I. Skolnik , INTRODUCTION TO RADAR SYSTEMS
[7 , Second Edition ,
McGraw-Hill , New York , 1980 , pp.78-79
] Tan Hsing Ho ; Shyh Jung Chung ; “A Compact Doppler Radar Using A New Hybrid Mixer ” , International Joint Conference of the 6th topic Symposium on [8
Millimeter Wave TSMMW2004
m; Jong-Hun Lee; Sang-Hyuk Kim; Sung-Ho Myung;
“A Novel Equivalent Circuit and Modeling Method for Defected Ground Structure and Its Application to Optimization of a DGS Low pass filter” , Microwave [9] Jun-Seok Park; Jae-Ho Ki
Symposium Digest, 2002 IEEE MTT-S International , Volume 1, 2-7 June 2002 Page(s): 417–420
[10] Ahn, D.; Park, Low-Pass Filte
J.-S.; Kim, C.-S.; Kim, J.; Qian, Y.; Itoh, T.; “A Design of the r Using the Novel Microstrip Defected Ground Structure” , Microwave Theory and Techniques , IEEE Transaction on , Volume 4, Issue 1,
[11] AVE TRANSISTOR AMPLIFIERS Analysis and
Jan. 2001 Page(s):86 – 93 Guillermo Gonzalez , MICROW
Design , Second Edition , Prentice Hall , 1997 , pp.385-386.
] Stephen A. Mass ,
[12 Nonlinear Microwave and RF Circuits , Second Edition , Artech House , 2003 , pp.542-545.
[13] Cicero S. Vaucher , Architectures for RF Frequency Synthesizers , Kluwer Academic ,2002 , pp.28-40.
[14] Stephen A. Mass , Nonlinear Microwave and RF Circuits , Second Edition , Artech House , 2003 , pp.477-481.
] Young-Taek Lee; Jong-Sik Lim
[15 ; Jun-Seok Park; Ahn, D.; Sangwook Nam; “ A
Novel Phase Noise Reduction Technique in Oscillator Using Defected Ground Structure ” , Microwave and Wireless Components Letters ,Volume 12, Issue 2, Feb. 2002 Page(s):39 - 41
] 張朝富,戴明
[16 鳳著;無線通信之射頻被動電路設計;全華科技圖,pp.6-28.