Professional Documents
Culture Documents
http://dreamcatcher.asia/cw
11
Introduction
An ideal filter is a linear 2-port network that provides
perfect transmission of signal for frequencies in a
certain passband region, infinite attenuation for
frequencies in the stopband region, and a linear
phase response in the passband (to reduce signal
distortion).
The goal of filter design is to approximate the ideal
requirements within acceptable tolerance with circuits
or systems consisting of real components.
Categorization of Filters
Low pass filter (LPF), high pass filter (HPF), bandpass filter (BPF),
bandstop filter (BSF), arbitrary type, etc.
In each category, the filter can be further divided into active and passive
types.
Active
Passive
Active
Passive
5
|H()|
Transfer
function
1
V1()
Filter
H()
V2()
ZL
V2
H
(1.1a)
V1
Complex value
Arg(H())
A()/dB
50
40
30
20
10
3
0
Real value
V2
Attenuation A 20 Log10
V1
c
(1.1b)
20log|s21()|
Zc
Filter
Zc
Arg(s21())
Transmission line
is optional
0 dB
b
b
s11 1
s21 2
a1 a 0
a1 a 0
2
2
Complex value
8
Passband
Transition band
Stopband
Filter
H()
V2()
ZL
|H()|
A()/dB
Transfer
function
Passband
50
40
30
20
10
3
0
Stopband
10
Bandstop filter
A()/dB
A()/dB
40
40
30
30
20
20
10
3
0
10
3
0
1 o 2
|H()|
1
1
|H()|
Transfer
function
Transfer
function
1 o 2
11
Filter
Zo
Zo
Zo
Zo
Zo
H1()
H2()
Consider a filter to be a
cascade of linear 2-port
networks.
Synthesize or realize each
2-port network, so that
the combine effect gives the
required frequency response.
The image impedance seen
at the input and output of
each network is maintained.
Zo
Response of
a single
network
Zo
Hn()
Zo
The combined
response
12
Filter
Zo
H s K
n 1
1
o
n
n
1
s bn 1s
b1s bo
Zo
13
Our Scope
Only concentrate on passive LC and stripline filters.
Filter synthesis using the Insertion Loss Method
(ILM). The Image Parameter Method (IPM) is more
efficient and suitable for simple filter designs, but has
the disadvantage that arbitrary frequency response
cannot be incorporated into the design.
14
15
16
tra
x
E
More on ILM
17
Vs
Lossless
2-port network
PA
Pin
ZL
PL
PLoad
2
2
1 1
PA 1 1
(2.1a)
PLR large,
high attenuation
P
LR large, high attenuation
P
closetoto1,1,low
lowattenuation
attenuation
PLRLRclose
For
Forexample,
example,aalow
lowpass
pass
filter
filterresponse
responseisisshown
shown
below:
below:
PLR(f)
High
attenuation
Low
attenuation
fc
f
18
PA
b2
Lossless
2-port network
Pin
Zc
PL
1a 2
2
1
PA
a1
2
PLR
PL
b2
1b 2
2
2
PLR 1
(2.1b)
2
s 21
19
(2.2)
M 2
1
N 2
1
2
1 1
M 2
1
M 2 N 2
PLR
M 2
M 2 N 2
PLR 1 P 2
M 2
N 2
(2.3a)
(2.3b)
function in P can
used for P(). The
P be
requirement is that P() must either be an odd or even function. Among the
classical polynomial functions are:
Maximally flat or Butterworth functions
Equal ripple or Chebyshev functions
Elliptic function
The
Thecharacteristics
characteristicswe
weneed
need
2
Many, many more
from [P()] 2for LPF:
20
P
c
(2.4a)
NN==order
orderofofthe
the
Characteristic
Characteristic
Polynomial
PolynomialP()
P()
P C N , ripple factor
C0 1
C N C1
C 2C C , n 2
n 1
n 2
n
(2.4b)
P 2 B j B j 1
B0 s 1
(2.4c)
BN s B1 s s 1
Bn s 2 s 1 Bn 1 s s 2 Bn 2 s , n 2
21
PLR of the low pass filter using 4th order polynomial functions (N
= 4) Butterworth, Chebyshev (ripple factor =1), and Bessel.
Normalized to c = 1 rad/s, k = 1.
2
2
PLR (chebyshev) 1 k 2 8 4 1
c
c
Ideal
1 10
1 10
k=1
Chebyshev
bt ( )
4
2
PLR ( Butterworth) 1 k
c
cb( )
PLR100
bs ( )
Butterworth
10
Bessel
1
0.5
1.5
PLR ( Bessel ) 1 k 2 B j B j 1
B s 1 s
105 c
10 s
c
45 s
c
105 s 105
c
22
PLR ( Butterworth) 1 k 2
c
5
1 10
PLR( 2) 4
1 10
N=7
PLR( 3)
PLR( 4)1 10
PLR( 5)
PLR( 6)
PLR( 7)
N=6
N=5
N=4
100
N=3
10
N=2
0.5
1.5
Conclusion:
Conclusion:
The
Thetype
typeofof
polynomial
polynomial
function
functionand
and
the
theorder
order
determine
determinethe
the
attenuation
attenuationrate
rate
ininthe
thestopband.
stopband.
23
24
Many researchers have tabulated the values for the L and C for the low
pass filter with cut-off frequency c = 1 rad/s, that works with the source
and load impedance Zs = ZL = 1 .
This low pass filter is known as the Low Pass Prototype (LPP).
L1=g2
C1=g1
L1=g1
g0= 1
L2=g4
C2=g3
RL= gN+1
L2=g3
C1=g2
C2=g4
RL= gN+1
Dual of each
other
25
26
g1
g2
g3
g4
g5
g6
g7
g8
1
2
3
4
5
6
2.0000
1.4142
1.0000
0.7654
0.6180
0.5176
1.0000
1.4142
2.0000
1.8478
1.6180
1.4142
1.0000
1.0000
1.8478
2.0000
1.9318
1.0000
0.7654
1.6180
1.9318
1.0000
0.6180
1.4142
1.0000
0.5176
1.0000
0.4450
1.2470
1.8019
2.0000
1.8019
1.2470
0.4450
1.0000
0.3902
1.1111
1.6629
1.9615
1.9615
1.6629
1.1111
0.3902
g9
1.0000
See Example 2.1 in the following slides on how the constant values g1, g2, g3,
etc., are obtained.
27
N
1
2
3
4
5
6
N
1
2
3
4
5
6
g2
1.0000
0.7071
1.0967
1.1926
1.2296
1.2479
g3
g4
g5
g6
g7
1.9841
1.5963
2.3661
2.5408
2.6064
1.0000
0.8419
1.2296
1.3137
1.9841
1.7058
2.4578
1.0000
0.8696
1.9841
g2
1.0000
0.5339
0.7117
0.7483
0.7618
0.7685
g3
g4
g5
g6
g7
5.8095
3.3487
4.3471
4.5381
4.6061
1.0000
0.5920
0.7618
0.7929
5.8095
3.4817
4.4641
1.0000
0.6033
5.8095
28
29
Vs
V1
R V
1 jRC s
R jL 1 jRRC
Thus
Vs
jL
R V1
1/jC
RV
RV
s
R R jL s1 jRC
2
2 R RLC j L R 2C
2
PL 21R V1
Vs R
and
2 2 2 LC R 2 2 L R 2C
PA 81R Vs
PLR P A
L
P
Vs
8R
Vs R
2
2
2 2 2 LC R 2 2 L R 2C
2
2 4
1 1 2 L R 2C LC 2 LC
2
4R
2
2
1 2 2 R 2 2 2 LC 2 L R 2C 2
8R
[P()]2
30
2
2 4
PLR 1 1 2 L R 2C LC 2 LC
1 a1 2 a2 4
2
4R
(E1.1)
2 1 4 1 0 2 1 4
PLR ( Butterworth) 1 2
(E1.3)
(E1.2)
L R 2C
2
a1 0 1 2 L R 2C LC 0
LC
4R
21R
(E1.4)
LC 2 C 2 2
LC 12 L C 2 L2 C 2 2 LC 0
C 2 1.4142
L C2 0
LC
Using (E1.3)
L C 1.4142
Compare
Comparethis
thisresult
resultwith
with
N=2
in
the
table
for
the
N=2 in the table for the
LPP
LPPButterworth
Butterworthresponse.
response.
This
direct
brute
force
This direct brute force
approach
approachcan
canbe
be
extended
to
N=3,
extended to N=3,4,4,5
5
31
tra
x
E
V_AC
SRC1
Vac=polar(1,0) V
Freq=freq
Vout
L
L1
L=1.4142 H
R=
C
C1
C=1.4142 F
R
R1
R=1 Ohm
32
tra
x
E
Eqn PL=0.5*mag(Vout)*mag(Vout)
Eqn PA=1/8
Eqn PLR=PA/PL
m1
m1
freq=160.0mHz
m1=-3.056
dB(Vout/0.5)
-10
2.5E4
PLR
2.0E4
1.5E4
-20
-30
-40
-50
1.0E4
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
freq, Hz
5.0E3
0.0
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
freq, Hz
33
34
LPP to
Low Pass
L
Zo L
c
LZ o
o
1
c LZ o
o LZ o
Zo
cC
C
Z o c
C
o Z o
Center frequency
o 1 2 2 or
1 2
1
o LZ o
Z o
oC
Fractional bandwidth
(2.5a)
LPP to
Bandstop
LPP to
Bandpass
2 1 (2.5b)
o
LZ o
o
Zo
o C
C
o Zo
Note
Notethat
thatthe
theinductor
inductoralways
always
multiplies
with
Z
while
the
o
multiplies with Zo while the
capacitor
capacitordivides
divideswith
withZZo
o
35
L1=g2
C1=g1
L2=g4
C2=g3
1
RL= gN+1
36
Vs
15.916pF
79.58nH
0.7072nH
0.1414pF
15.916pF
1
0.7072nH
RL
50
All
Alluses
usesthe
themicrostrip
microstripstripline
striplinecircuit
circuit
37
L1=0.7654H
g 0= 1
L2=1.8478H
C1=1.8478F
L1=4.061 nH
g0=1/50
C2=0.7654F
RL= 1
L
L Zo n
c
Cn
C
Z o c
L2=9.803 nH
C1=3.921 pF
C2=1.624 pF
R Z o Rn
RL= 50
39
L1=1.6703H
g 0= 1
L2=2.3661H
C1=1.1926F
g0=1/50
C2=0.8419F
RL=
1.9841
L
L Zo n
c
Cn
C
Z o c
L2=12.55 nH
C1=2.531 pF
C2=1.787 pF
R Z o Rn
RL=
99.2
40
dB(LPF_butterworth..S(2,1))
dB(S(2,1))
5
0
Butterworth
-10
|s21|
-20
Chebyshev
-30
0.5
1.0
1.5
2.0
2.5
3.0
freq, GHz
-50
P hase_butterworth
P hase_chebyshev
0.0
Better phase
linearity for Butterworth
LPF in the passband
-100
-150
Butterworth
Arg(s21)
-200
-250
Chebyshev
-300
-350
0.0
0.5
1.0
1.5
2.0
2.5
3.0
freq, GHz
41
42
From table, design the low pass prototype (LPP) for 3rd order
Butterworth response, c = 1.
Zo=1
2<0o
g2
2.000H
g1
1.000F
g3
1.000F
g4
1
Simulated result
using PSpice
c 2f c 1
f c 21 0.1592 Hz
Voltage across g4
43
Impedance denormalization
1 2 1.4GHz
2 2 1.6GHz
fo
o LZ o
50
Vs
f1 f 2 1.497GHz
2
o
C
o Z o
79.58 nH
0.133
Z o
oC
0.1414 pF
RL
15.916 pF
0.7072 nH
15.916 pF
0.7072 nH
50
44
Voltage across RL
45
tra
x
E
There is also another class of filter known as the All Pass Filter (APF).
This type of filter does not produce any attenuation in the magnitude
response, but provides phase response in the band of interest.
|H(f)|
Arg(H(f))
1
0
Nonlinear
phase in
passband
f
|H(f)|
Arg(H(f))
1
0
|H(f)|
BPF
APF
Zo
Arg(H(f))
1
0
Linear
phase in
passband
f
46
C
Ct3
C=Ct_value2 pF
L
Lt1
L=Lt_value nH
R=
Term
Term1
Num=1
Z=50 Ohm
C
Ct1
C=Ct_value pF
CP WSUB
CP WSub1
H=62.0 mil
Er=4.6
Mur=1
Cond=5.8E+7
T=1.38 mil
TanD=0.02
Rough=0.0 mil
S-PARAMETERS
Var
E qn
S_P aram
SP 1
Start=0.1 GHz
Stop=3.0 GHz
Step=1.0 MHz
INDQ
1_0pF_NP O_0603
CP WG
L4
C1
CP W1
L=15.0 nH
Q=90.0
Subst="CP WSub1"
b82496c3229j000
4_7pF_NPF=
O_0603
800.0 MHz
W=50.0 mil
L2
C2
G=10.0 mil
Mode=proportional to freq
param=
SIMID
0603-C
(2.2
nH
+
-5%)
L=28.0 mm
Rdc=0.1 Ohm
VAR
VAR1
Lt_value=4.8
Ct_value=3.5
Ct_value2=2.9
C
Ct2
C=Ct_value pF
CP WG
CP W2
b82496c3229j000
Subst="CP WSub1"
L3
W=50.0
mil
4_7pF_NP
O_0603
param=SIMID 0603-C (2.2
nH +-5%)
G=10.0 mil
C3
L=28.0 mm
L
Lt2
L=Lt_value nH
R=
C
Ct45
C=Ct_value2 pF
T erm
T erm2
Num=2
Z=50 Ohm
47
48
Measured
Simulated
-20
-20
-40
-40
Arg(s21)/degree
-60
-60
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0
freq, GHz
freq, GHz
200
200
phase(RF_BPF_measured..S(2,1))
phase(RF_BPF_m
easure
d))
..S(2,1))
phase(S(2
,1
phase(S(2,1))
dB(RF_BPF_measured..S(2,1))
dB(RF_BPF_m
eB(S(2
asure,1
d..S(2
d
)) ,1))
dB(S(2,1))
100
100
-100
-100
-200
-200
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0
freq, GHz
freq, GHz
49
50
51
52
L
Zo
Zc ,
Zc ,
Zo
Zc ,
Zo
Zo
53
Zc ,
Zc ,
Zc ,
Thus
Thussome
sometheorems
theoremsare
areused
usedtoto
facilitate
facilitatethe
thetransformation
transformationof
ofthe
theLC
LC
circuit
circuitinto
intostripline
striplinemicrowave
microwavecircuits.
circuits.
Chief
Chiefamong
amongthese
theseare
arethe
theKurodas
Kurodas
Identities
Identities(See
(SeeAppendix
Appendix1)
1)
54
Zin
Zc L
Zc ,
(3.1.1a)
Zin
Zc ,
tan l c 1
Zc
Wavelength at
cut-off frequency
tan
l 1 l
(3.1.1c)
8
c
55
g4
1
g2
2.000F
Zc=1.000
Zo=1
g1
1.000H
Zc=0.500
1 0.500
2.000
Zc=1.000
56
Z1 1.0
Z2=1
Yc
1
n2Z
0.5
2
n2Z1=2
Zc=1.000
Extra Tline
Zc=1.000
Extra Tline
Z
n2 1 2
Z1
1 1 2
1
Zc=1.0
Zc=1.0
Zc=0.500
Similar operation is
performed here
Length = c/8
for all Tlines
at = 1 rad/s
57
Zc=2.0
Zc=2.000
Zc=0.500
Zc=2.000
Length = c/8
for all Tlines
at = 1 rad/s
Since
Sinceall
allTlines
Tlineshave
havesimilar
similarphysical
physical
length,
length,this
thisapproach
approachto
tostripline
striplinefilter
filter
implementation
implementationisisalso
alsoknown
knownas
as
Commensurate
CommensurateLine
LineApproach.
Approach.
58
Zc=100
Zc=100
Zc=25
Zc=100
Length = c/8
for all Tlines at
f = fc = 1.5 GHz
50
/8 @ 1.5 GHz/mm
13.45
12.77
14.23
W/mm
2.85
8.00
0.61
We
Wecan
canwork
workout
outthe
thecorrect
correctwidth
widthW
Wgiven
giventhe
the
impedance,
dielectric
constant,
and
thickness.
impedance, dielectric constant, and thickness.
From
FromW/H
W/Hratio,
ratio,the
theeffective
effectivedielectric
dielectricconstant
constant
eff can
be
determined.
Use
this
together
with
eff can be determined. Use this together with
frequency
frequencyatat1.5
1.5GHz
GHztotofind
findthe
thewavelength.
wavelength.
59
60
MSub
MTEE
Tee1
Subst="MSub1"
W1=2.85 mm
W2=0.61 mm
W3=0.61 mm
MLIN
TL1
Subst="MSub1"
W=2.85 mm
L=25.0 mm
Term
Term1
Num=1
Z=50 Ohm
MTEE
Tee3
Subst="MSub1"
W1=0.61 mm
W2=0.61 mm
W3=8.00 mm
MLIN
TL3
Subst="MSub1"
W=0.61 mm
L=14.23 mm
MLOC
TL6
Subst="MSub1"
W=0.61 mm
L=14.23 mm
MTEE
Tee2
Subst="MSub1"
W1=0.61 mm
W2=2.85 mm
W3=0.61 mm
MLIN
TL4
Subst="MSub1"
W=0.61 mm
L=14.23 mm
MLOC
TL5
Subst="MSub1"
W=8.0 mm
L=12.77 mm
MLIN
TL2
Subst="MSub1"
W=2.85 mm
L=25.0 mm
MLOC
TL7
Subst="MSub1"
W=0.61 mm
L=14.23 mm
L
L2
C
L=5.305 nH
C1
R=
C=4.244 pF
Term
Term2
Num=2
Z=50 Ohm
m1
freq=1.500GHz
m1=-6.092
Term
Term2
Num=2
Z=50 Ohm
dB(Butter_LPF_LC..S(2,1))
dB(S(2,1))
MSUB
MSub1
H=1.57 mm
Er=4.6
Mur=1
Cond=1.0E+50
Hu=3.9e+034 mil
T=0.036 mm
TanD=0.02
Rough=0 mil
Term
Term1
Num=1
Z=50 Ohm
S_Param
SP1
Start=0.2 GHz
Stop=4.0 GHz
Step=5 MHz
L
L1
L=5.305 nH
R=
m1
-10
-20
-30
-40
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
freq, GHz
61
62
63
64
Z11 Z12
Z12
Positive
susceptance
Z11 Z12 jZ c
jZ c
1
sin l
jZ c
sin 2
l
2
l
2
1 cos 2
jZ c tan
l
cos
sin l
2 sin 2 cos 2
l
2
Positive reactance
2 sin 2 2
l
2
l
2
l
2
Zc
Z11 Z 22 jZ c cot l
(3.2.1a)
(3.2.1c)
65
When l < /2, the series element can be thought of as an inductor and
the shunt element can be considered a capacitor.
X
l
Z11 Z12 Z c tan
2
2
1
1
B
sin l
Z12
Zc
X ZH l
B0
1
X 0
B
l
ZL
Z11 - Z12
When Zc >> 1
l < /4
Z12
jX/2
X ZH l
jX/2
jB
When Zc 1
l < /4
B YLl
66
lL
c L
ZH
c CZ L
lC
(3.2.2a)
(3.2.2b)
67
Let us use the microstrip line. Since a microstrip Tline with low
Zc is wide and a Tline with high Zc is narrow, the transformation
from circuit to physical layout would be as follows:
L1=4.061 nH
g0=1/50
L2=9.803 nH
C1=3.921 pF
C2=1.624 pF
RL= 50
68
Zc = 15
Zc = 50
Zc = 110
W/d
10.0
2.0
0.36
d/mm
1.5
1.5
1.5
W/mm
15.0
3.0
0.6
e
3.68
3.21
2.83
L eL ko eL 2f c 3.3356 10 9 60.307 s 1
H eH ko eH 2f c 3.3356 10 9 53.258s 1
69
c L1
6.5mm
ZH H
Verification:
H l1 0.392 4 0.7854
L l2 0.490 4 0.7854
CZ
l2 c 1 L 9.2mm
L
H l3 0.905 4 0.7854
l3 15.0mm
l4 3.8mm
3.0 mm
L l4 0.202 4 0.7854
l1
l2
l3
Nevertheless we still
proceed with the implementation. It will be seen
that this will affect the
accuracy of the 3 dB cut-off
point of the filter.
l4
50 line
50 line
To 50
Load
15.0 mm
0.6 mm
70
71
m1
dB(S(2,1))
-5
-10
-15
-20
-25
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
freq, GHz
72
m1
-5
dB(S(2,1))
-15
-25
0
freq, GHz
73
74
Vs
79.58 nH
15.916 pF
To source
network
o
4
J1
90o
0.7072 nH
Admittance
inverter
Tline
J2
90o
0.1414 pF
15.916 pF
J4
90o
To RL
An array of coupled
microstrip line
o
4
Section 1
0.7072 nH
RL
50
An equivalent circuit
model for coupled Tlines
with open circuit at
two ends.
Section 2
Section 3
Section 4
o ==wavelength
wavelengthatatoo
o
75
Zoo and Zoe are in turn depend on the gain of the corresponding
admittance inverter J.
From Example 2.3
Z oe Z o 1 JZ o JZ o 2
fo
Z oo
f1 f 2 1.497GHz
2
o
2 g1
J1 Z1
Z 1 JZ
1 2 1.4GHz
2 2 1.6GHz
0.133
JZ
o
o
J n 2 Z1
for n 2,3,4 N
g n 1g n
o
J N 1 Z1
2 g N g N 1
76
Section 1:
J1
0.009163
2 g1
1
Zo
Section 2:
J2
1
2Zo
Z oo 2 Z o 1 J 2 Z o J 2 Z o 2 43.680
0.002969
g1g 2
Z oe 2 Z o 1 J 2 Z o J 2 Z o 2 58.523
0.002969
g 2 g3
Z oe3 83.403
0.009163
2 g3 g 4
Z oe 4 58.523
Section 3:
J 3 2 Z1
Z oo3 37.588
Section 4:
J 4 Z1
o
Z oo 4 43.680
Note:
Note:
gg1=1.0000
1=1.0000
gg2=2.0000
2=2.0000
gg3=1.0000
3=1.0000
gg4=1.0000
=1.0000
4
77
78
79
80
Strategy:
1) We tune the W and
S for the specified Zoo and Zoe.
2) Based on the width W of
a single trace, we work out
the effective permittivity, and
use this to calculate the phase
velocity.
3) From this we find the
wavelength at 1.5GHz and
work out the required quarter
wavelength.
81
82
mag(S(2,1))
0.8
0.6
0.4
0.2
0.0
1.0
1.5
2.0
2.5
2fo
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
9.5
10.0
freq, GHz
83
84
mag(S(2,1))
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
9.5
10.0
freq, GHz
85
86
LPF
HPF:
BPF:
SMD capacitor
87
More BPFs:
BSF:
88
89
Kurodas Identities
As extracted from
Ref. [2]
l
1
Z2
Z1
Z2
Z1
Z1
n
n2Z1
1
2
n Z2
Z2/n
Note:
Note:The
The
length
lengthofofall
all
transmission
transmission
lines
linesisis
l l==/8
/8
1: n2
2
Z1
n2
l
1
Z2
l
Z1
Z2
Z1
Z2/n
l
Z1
Z
n2 1 2
n2Z1
n2: 1
1
n2Z
90
References
[1] R. E. Collin, Foundations for Microwave Engineering, 2nd Edition 1992,
McGraw-Hill.
[2] D. M. Pozar, Microwave Engineering, 2nd Edition 1998, John-Wiley &
Sons.* (3rd Edition 2005, John-Wiley & Sons is now available)
Other more advanced references:
[3] W. Chen (Editor), The Circuits and Filters Handbook, 1995, CRC
Press.*
[4] I. Hunter, Theory and Design of Microwave Filters, 2001, The Institution
of Electrical Engineers.*
[5] G. Matthaei, L. Young, E.M.T. Jones, Microwave Filters, ImpedanceMatching Networks, and Coupling Structures, 1980, Artech House.*
[6] F. F. Kuo, Network Analysis and Synthesis, 2nd Edition 1966, JohnWiley & Sons.
**Recommended
Recommended
91