CHAPTER 2 CIRCUIT DESIGN AND SIMULATION
2.3 Simulation Results
The post-simulation is completed by ADS simulator for whole simulation with process parameters of TSMC 0.13-um CMOS MS/RF general purpose 1P8M salicide Cu-FSG 1.2V/2.5V RF SPICE models. The following diagrams are the post-simulation results of thorough receiver circuits.
LNA
Because the first stage of receiver is LNA, it must provide input matching, power gain and low-noise contribution in a specific frequency band. The figures from Fig. 34 to Fig. 37 are the port-simulation results of S-parameters, S11, S22, S21 and S12, respectively. The input and output matching are lower than -10dB indicated a power transfer greater than 90%. Fig. 38 exhibits that the noise figure is lower 3.5dB at the frequency of 24-GHz. The resultant noise figure is close to the minimum noise figure.
The simulation results of linearity are shown in Fig. 39, Fig. 40 and Fig. 41. The linearity performance of input 1-dB compression (P1dB) is about -21.3dBm. The input third-intercept point (IIP3) is about -11.7dBm. The transient simulation results of input and output current amplitude are shown in Fig. 42. The stability simulation results of stability factor and stability measure are shown in Fig. 43 and Fig. 44. The necessary and sufficient conditions for unconditional stability are that the stability factor is greater then unit and the stability measure are positive. The Table(iii) is the comparison with previously reported LNA for frequencies around 20-GHz. The conversion gain of current-mode LNA is dominant with the device size ratio of current-mirror amplifier.
Thus, the supply voltage has slight effect on the conversion gain. But the supply voltage would have great effects on the power consumption.
20 22 24 26 28
18 30
-10 -8 -6 -4
-12 -2
freq, GHz
dB(S(1,1))
m3
m3 freq=
dB(S(1,1))=-10.526 24.00GHz
Fig. 34 Simulation result S11 of LNA.
20 22 24 26 28
18 30
-30 -20 -10
-40 0
freq, GHz
dB(S(2,2))
m4
m4 freq=
dB(S(2,2))=-24.477 24.00GHz
Fig. 35 Simulation result S22 of LNA.
20 22 24 26 28
18 30
5 10 15
0 20
freq, GHz
dB (S (2 ,1 ))
m5
m5 freq=
dB(S(2,1))=17.117 24.00GHz
Fig. 36 Simulation result S21 of LNA.
20 22 24 26 28
18 30
-45 -40 -35
-50 -30
freq, GHz
dB(S(1,2))
m8 m8 freq=
dB(S(1,2))=-46.130 24.00GHz
Fig. 37 Simulation result S12 of LNA.
20 22 24 26 28
18 30
4 5 6 7
3 8
freq, GHz
nf(2)
m6
NFmin
m7 m6 freq=
nf(2)=3.447 24.00GHz m7 freq=
NFmin=3.391 24.00GHz
Fig. 38 Simulation result NFmin of LNA.
-35 -30 -25 -20 -15 -10 -5
-40 0
-20 -10 0 10
-30 20
RF_pwr
dBm(HB1.HB.Vo_LNA[::,1])
m1
P1dB m2
m1RF_pwr=
dBm(HB1.HB.Vo_LNA[::,1])=-22.875-40.000 m2
RF_pwr=
P1dB=-5.175-21.300
Fig. 39 Linearity simulation result P1dB of LNA.
-35 -30 -25 -20 -15 -10 -5
-40 0
0 5 10 15
-5 20
RF_pwr
Conversion_gain
m11 m12
m11 RF_pwr=
Conversion_gain=17.125 -40.000 m12
RF_pwr=
Conversion_gain=16.114 -21.300
Fig. 40 Simulation result of conversion gain of LNA.
-40 -30 -20 -10
-50 0
-100 -50 0
-150 50
RF_pwr
dBm(mix(HB.Vo_LNA,{2,-1}))
m2
dBm(mix(HB.Vo_LNA,{1,0}))
m1
first_hormonicthird_hormonic
m2RF_pwr=
dBm(mix(HB.Vo_LNA,{2,-1}))=-109.527-50.000 m1RF_pwr=
dBm(mix(HB.Vo_LNA,{1,0}))=-32.948-50.000
RF_pwr -50.000
IIP3 -11.711
Fig. 41 Simulation result IIP3 of LNA.
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
0.0 2.0
-100 0 100
-200 200
time, nsec
TRAN.I_LNAout.i, uA
m3
TRAN.I_LNAin.i, uA
m7 m3 time=
TRAN.I_LNAout.i=162.0uA 620.4psec m7 time=
TRAN.I_LNAin.i=25.84uA 615.2psec
Fig. 42 Transient simulation result of LNA.
Eqn K=stab_fact(S)
20 22 24 26 28
18 30
4 6 8 10 12
2 14
freq, GHz
K
Fig. 43 Simulation result K-factor of LNA.
20 22 24 26 28
18 30
0.4 0.6 0.8 1.0 1.2
0.2 1.4
freq, GHz
b
Eqn b = stab_meas(S)
Fig. 44 Simulation result stability measure of LNA.
0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3
0 10 20 30 40 50 60
0 10 20 30 40 50 60
Power consumption [mW]
S21 [dB]
Supply voltage of current-mode LNA [V]
S21
power consumption
Fig. 45 The S21 and power consumption of LNA in the different supply voltage.
Table(iii) The comparison with previously reported LNA for frequencies around 20-GHz.
Reference This work This work [8] [21] [7] [22]
Freq. [GHz] 24 24 21.8 20 23.7 24
Technology 0.13um 0.13um 0.18um 0.13um 0.18um 0.18um
Topology
2 stage
Current-mode
2 stage
Current-mode
3 stage
CGRF+CS+CS
2 stage
CS+CS
3 stage
CS+CS+CS
2 stage
CS+CS
Gain [dB] 17.1 17.5 15 12.9 12.86 13.1
P1dB [dBm] -21.3 -19.7 -- -- -- --
IIP3 [dBm] -11.7 -10.8 -- -- -- --
NF [dB] 3.4 3.5 6.0 4.4 5.6 3.9
S11 [dB] <-10 <-10 -21 -14 -11 -15
Total power [mW]
10.6
@1.2V
8.8
@0.7V
24
@1.5V
16.8
@1.2V
54
@1.8V
14
@1V
overall
The Fig. 46, Fig. 47 and Fig. 48 plot the S-parameter simulation results at RF, LO and output port in the whole receiver, respectively. The harmonic simulation is shown in Fig. 49. The input spectrum at RF (24-GHz) and LO port (19-GHz) are about -49dBm and -3dBm, respectively. The output spectrum of receiver at 5-GHz is about -27.898dBm. Thus, the conversion gain of overall receiver at 24-GHz is about 20.913dB. From the Fig. 49(c), some of sub-harmonic the isolation performance can be found. The spectrum at LO frequency (19-GHz), RF frequency (24-GHz), up-conversion frequency (19-GHz+24-GHz), the double frequency of RF and LO frequency (48-GHz and 38-GHz) and third frequency of RF and LO frequency
(72-GHz and 57-GHz) can be suppressed with resonating at 5-GHz between Lsq1 and Csq2. The conversion gain and noise figure versus frequency of overall receiver path are shown in Fig. 50 and Fig. 51. The frequency range of K-band applications is from 18GHz to 26.5GHz. The frequency bandwidth of ISM band is about 1GHz. The 3-dB bandwidth of current-mode receiver is about 1.75 GHz (21.5GHz~23.25 GHz) and conversion gain of current-mode receiver is about 26.4 dB at the peak frequency of 22.5GHz. The 3-dB bandwidth is dominant with the Q value of inductor. For the higher conversion gain, the higher Q of the inductor will be used. Therefore, the wider 3dB bandwidth and higher conversion gain are trade-off. Finally, the diagram of Fig. 52 and Fig. 53 present the simulation results of linearity performance. The summary table of post-simulation with corner variations presents as Table(iv). The Table(v) is a summary table of post-simulation with temperature variation.
5 10 15 20 25
0 30
-10 -8 -6 -4 -2
-12 0
freq, GHz
dB (S (1, 1 ))
m5 m5 freq=
dB(S(1,1))=-10.525 24.00GHz
Fig. 46 S-parameter simulation result at RF input of the overall receiver.
5 10 15 20 25
0 30
-25 -20 -15 -10 -5
-30 0
freq, GHz
dB (S (2, 2 ))
m6 m6 freq=
dB(S(2,2))=-26.720 19.00GHz
Fig. 47 S-parameter simulation result at LO port of the overall receiver.
5 10 15 20 25
0 30
-15 -10 -5
-20 0
freq, GHz
dB (S (3, 3 ))
m14
m14 freq=
dB(S(3,3))=-17.283 5.000GHz
Fig. 48 S-parameter simulation result at output of the overall receiver.
20 40 60 80 100 120
0 140
-200 -150 -100 -50
-250 0
freq, GHz dBm(HB1.HB.Vin_LNA)
m1 m1
freq=
dBm(HB1.HB.Vin_LNA)=-48.811 24.00GHz
(a)
20 40 60 80 100 120
0 140
-150 -100 -50
-200 0
freq, GHz
dBm(HB1.HB.Vin_LO)
m8
m8 freq=
dBm(HB1.HB.Vin_LO)=-2.933 19.00GHz
(b)
20 40 60 80 100 120
0 140
-140 -120 -100 -80 -60 -40
-160 -20
freq, GHz dBm(HB1.HB.Vout_square) m7 m9 m10
m18m17 m7freq=
dBm(HB1.HB.Vout_square)=-27.8985.000GHz m9 freq=
dBm(HB1.HB.Vout_square)=-33.26019.00GHz m10freq=
dBm(HB1.HB.Vout_square)=-40.84938.00GHz
m17freq=
dBm(HB1.HB.Vout_square)=-63.84257.00GHz m18freq=
dBm(HB1.HB.Vout_square)=-71.99343.00GHz
(c)
Fig. 49 The harmonic simulation results of the receiver. (a) The spectrum diagram at RF input port. (b) The spectrum diagram at LO port. (c) The spectrum diagram at output port.
20 21 22 23 24 25 26 27 28
10 12 14 16 18 20 22 24 26 28
Gain of the overall receiver (dB)
Frequency (GHz)
B
Fig. 50 The conversion gain of the overall receiver.
20 21 22 23 24 25 26 27 28 4.0
4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
Noise figure (dB)
Frequency (GHz)
C
Fig. 51 The noise figure of the overall receiver.
-35 -30 -25 -20
-40 -15
-15 -10 -5 0 5
-20 10
RF_pwr
dBm(HB2.HB.Vout_square[::,1])
m15
P1dB_Vo m11
m15RF_pwr=
dBm(HB2.HB.Vout_square[::,1])=-18.095-40.000 m11 RF_pwr=
P1dB_Vo=-3.595-24.500
Fig. 52 The simulation result P1dB of the overall receiver.
-35 -30 -25 -20
-40 -15
16 18 20
14 22
RF_pwr
Conversion_gain
m19
m20 m19 RF_pwr=
Conversion_gain=21.905 -40.000 m20
RF_pwr=
Conversion_gain=20.900 -24.400
Fig. 53 The linearity simulation result of overall receiver.
Table(iv) The corner of the post-simulation summary.
FF TT SS
GainLNA [dB] 18.5 17.1 13.9
NFLNA [dB] 3.5 3.4 3.6
GainRX [dB] 24.7 20.9 14.5
NFRX [dB] 4.1 4.2 5.2
P1dB [dBm] -26.2 -24.4 -25.7
Power dissipation [mW] 37.4 24.3 14.7
S11 (<10 dB) 23.5~25.5 GHz 23.2~24.3 GHz 22.9~23.6 GHz S22 (<10 dB) >14.8 GHz >13.5 GHz >10.1 GHz S33 (<10 dB) 4.2~7.0 GHz 4.0~6.3 GHz 3.5~5.9 GHz
Table(v) Post-simulation summary with temperature variation.
Temperature 0 25 100
GainLNA [dB] 16.7 17.1 17.4
NFLNA [dB] 3.2 3.4 4.0
GainRX [dB] 21.7 20.9 18.0
NFRX [dB] 4.0 4.2 4.9
P1dB [dBm] -23.8 -24.4 -26.9
Power dissipation [mW] 23.6 24.3 26.2
S11 (<10 dB) 23.2~24.1 GHz 23.2~24.3 GHz 23.2~24.8 GHz S22 (<10 dB) >13.9 GHz >13.5 GHz >12.5 GHz S33 (<10 dB) 3.9~6.3 GHz 4.0~6.3 GHz 3.8~6.4 GHz