EN4520-Microwave Communications

Microwave Network Analysis

Department of Electronic and Telecommunication Engineering University of Moratuwa

University of Moratuwa

1 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Introduction

Introduction
A microwave (MW) network is formed when several MV devices and

components such as sources, attenuators, resonators, lters, ampliers, etc. are coupled together by transmission lines or waveguides for the desired transmission of a MW signal.
The point of interconnection of two or more devices is called a junction. RF/MW devices, circuits, and components can be classied as one-, two-,

three-, or N-point networks. A majority of circuits under analysis are two-port networks.
We can use network and/or transmission line theory to analyze the

behavior of the entire system of components, including eects such as multiple reections, loss, impedance transformations, and transitions. A transition between dierent transmission lines, or discontinuity on a transmission line generally can not be treated as a simple junction between two transmission lines.

University of Moratuwa

2 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Impedance and Equivalent Voltages and Currents

Equivalent Voltages and Currents

At microwave frequencies the measurement of voltage or current is

dicult (or impossible), unless a clearly dened terminal pair is available.

Such a terminal pair may be present in the case of TEM-type lines (such

as coaxial cable, microstrip, or stripline), but does not strictly exist for non-TEM lines (such as rectangular, circular, or surface waveguides). There are many ways to dene equivalent voltage, current, and impedance for waveguides, since these quantities are not unique for non-TEM lines:
Voltage and current are dened only for a particular waveguide mode, and

are dened so that the voltage is proportional to the transverse electric eld, and the current is proportional to the transverse magnetic eld. In order to be used in a manner similar to voltages and currents of circuit theory the equivalent voltages and currents should be dened so that their product gives the power ow of the mode. The ratio of the voltage to the current for a single traveling wave should be equal to the characteristic impedance of the line.

University of Moratuwa

3 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Impedance and Admittance Matrices

Impedance and Admittance Matrices

Once equivalent voltages and currents have been dened at various points

in a microwave network, we can use the impedance and/or admittance matrices of circuit theory to relate these terminal or port quantities to each other. This type of representation lends itself to the development of equivalent circuits of arbitrary networks. Consider an arbitrary N -port microwave network.
The ports may be any type of transmission line or transmission line

equivalent of a single propagating waveguide mode.

If one of the physical ports of the network is a waveguide supporting more

than one propagating mode, additional electrical ports can be added to account for these modes. At a specic point on the nth port, a terminal plane, tn , is dened along + + with equivalent voltages and currents for the incident (Vn , In ) and reected (Vn , In ) waves.
University of Moratuwa
4 / 35 EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Impedance and Admittance Matrices

Figure: An arbitrary N -port microwave network. The terminal planes are important in providing a phase reference for the

voltage and current phasors.

University of Moratuwa
5 / 35 EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Impedance and Admittance Matrices

At the nth terminal plane, the total voltage and current is given by

Vn In when z = 0.

= =

+ Vn + Vn , + In In ,

The impedance matrix [Z] of the microwave network then relates these

voltages and currents: V1 V2 . . . VN Z21 = . . . ZN 1 Z11 Z12 Z1N . . . . . . ZN N I1 I2 . . . IN ,

or in matrix form as [V ] = [Z][I].

University of Moratuwa

6 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Impedance and Admittance Matrices

Similarly, we can dene an admittance matrix [Y ] as

I1 I2 . . . IN

Y21 = . . . YN 1

Y11

Y12

Y1N . . . . . . YN N

V1 V2 . . . VN

or in matrix form as [I] = [Y ][V ].

The [Z] and [Y ] matrices are the inverses of each other:

[Y ] = [Z]1 .
Zij can be found as

Zij =

Vi |I =0 Ij k

for k=j

University of Moratuwa

7 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Impedance and Admittance Matrices

Zij can be found by driving port j with the current Ij , open-circuiting all

other ports (so Ik = 0 for k = j), and measuring the open-circuit voltage at port i. Thus, Zii is the input impedance seen looking into port i when all other ports are open-circuited, and Zij is the transfer impedance between ports i and j when all other ports are open-circuited. Yii can be found as Ii Yij = |V =0 for k=j Vj k which states that Yij can be determined by driving port j with the voltage Vj , short-circuiting all other ports (so Vk = 0 for k = j), and measuring the short-circuit current at port i. In general, each Zij or Yij element may be complex.

For an arbitrary N -port network, the impedance and admittance matrices

are N N in size, so there are 2N 2 independent quantities or degrees of freedom. In practice, however, many networks are either reciprocal or lossless, or both.
University of Moratuwa
8 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Impedance and Admittance Matrices

Reciprocal Networks
If the network is reciprocal, the impedance and admittance matrices are

symmetric, so that Zij = Zji , and Yij = Yji .

Either of these special cases serves to reduce the number of independent

quantities or degrees of freedom that an N -port network may have. Consider the arbitrary network to be reciprocal (no active devices, ferrites, or plasmas), with short circuits placed at all terminal planes except those of ports 1 and 2. Now let Ea , Ha and Eb , Hb , be the elds anywhere in the network due to two independent sources, a and b, located somewhere in the network.
Then the reciprocity theorem states that

s Ea Hb .d =
S S

s Eb Ha .d,

where we will take S as the closed surface along the boundaries of the network and through the terminal planes of the ports.
University of Moratuwa
9 / 35 EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Impedance and Admittance Matrices

The elds due to sources a and b can be evaluated at the terminal planes

t1 and t2 as E1a E1b E2a E2b = V1a e1 = V1b e1 = V2a e2 = V2b e2 H1a = I1a h1 1b = I1b h1 H H2a = I2a h2 H2b = I2b h2 ,

where e1 , h1 and e2 , h2 are the transverse modal elds of ports 1 and 2, respectively, and the V s and Is are the equivalent total voltages and currents. Substituting the elds gives s s (V1a I1b V1b I1a ) e1 h1 .d + (V2a I2b V2b I2a ) e2 h2 .d = 0, (1)
S1 S2

where S1 , S2 are the cross-sectional areas at the terminal planes of ports 1 and 2.
University of Moratuwa
EN4520-Microwave Communications Chandika Wavegedara

10 / 35

Microwave Network Analysis

Impedance and Admittance Matrices

The equivalent voltages and currents have been dened so that the power

through a given port can be expressed as V I /2; We have C1 = C2 = 1 for each port, so that 1 .d = s e1 h s e2 h2 .d = 1.
S1 S2

This reduces (1) to

V1a I1b V1b I1a + V2a I2b V2b I2a = 0. to eliminate the Is: I1 I2
Substitution into (2) gives

(2)

Now use the 2 2 admittance matrix of the (eectively) two-port network

= Y11 V1 + Y12 V2 , = Y21 V1 + Y21 V2 .

(V1a V2a V1b V2a )(Y12 Y21 ) = 0.

University of Moratuwa
11 / 35

(3)

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Impedance and Admittance Matrices

Since the sources a and b are independent, the voltages V1a , V1b , V2a , and

V2b can take on arbitrary values.

So in order for (3) to be satised for any choice of sources, we must have

Y12 = Y21 , and we have the general result that Yij = Yji .
Then if [Y ] is a symmetric matrix, its inverse, [Z], is also symmetric.

University of Moratuwa

12 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Impedance and Admittance Matrices

Lossless Networks
Now consider a reciprocal lossless N -port junction. If the network is lossless, then the net real power delivered to the network

must be zero. Thus, Re{Pav } = 0, where Pav = =

1 t 1 [V ] [I] = [I]t [Z][I] 2 2 N N 1 Im Zmn In . 2 n=1 m=1

Since the In s are independent, we must have the real part of each self

term (In Znn In ) equal to zero, since we could set all port currents equal to zero except for the nth current. So, Re{In Znn In } = |In |2 Re{Znn } = 0, or Re{Znn } = 0.

(4)

University of Moratuwa

13 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Impedance and Admittance Matrices

Now let all port currents be zero except for Im and In . Then,
Re{In Im + Im In } = 0,

(5)

Since Zmn = Znm . But (In Im + Im In ) is a purely real quantity which is, in general, nonzero. Thus we must have that Re{Zmn } = 0. Then (4) and (5) imply that Re{Zmn } = 0 for any m, n, i.e., the elements of the impedance and admittance matrices must be pure imaginary.

University of Moratuwa

14 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Scattering Matrix

Scattering Matrix
A practical problem exists when trying to measure voltages and currents

at microwave frequencies because direct measurements usually involve the magnitude and phase of a wave traveling in a given direction, or of a standing wave. Thus, equivalent voltages and currents, and the related impedance and admittance matrices, become somewhat of an abstraction when dealing with high-frequency networks.
A representation more in accord with direct measurements and with the

ideas of incident, reected, and transmitted waves, is given by the scattering matrix. The scattering matrix provides a complete description of the network as seen at its N ports. The scattering matrix relates the voltage waves incident on the ports to those reected from the ports. For some components and circuits, the scattering parameters can be calculated using network analysis techniques.
University of Moratuwa
15 / 35 EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Scattering Matrix

Otherwise, the scattering parameters can be measured directly with a

vector network analyzer.

Figure: Helwlet-Packard HP8510B Network Analyzer

University of Moratuwa
EN4520-Microwave Communications Chandika Wavegedara

16 / 35

Microwave Network Analysis

Scattering Matrix

Once the scattering parameters of the network are known, conversion to

other matrix parameters can be performed, if needed. + Consider the N -port network shown in the gure, where Vn is the amplitude of the voltage wave incident on port n, and Vn is the amplitude of the voltage wave reected from port n. The scattering matrix, or [S] matrix, is dened in relation to these incident and reected voltage waves as S11 S12 S1N V1 V+ 1+ . V2 . V2 S11 . . = . . . , . . . . . . . . + VN VN SN 1 SN N or [V ] = [S][V + ].

University of Moratuwa

17 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Scattering Matrix

A specic element of the [S] matrix can be determined as

Sij =

Vi | + Vj+ Vk =0

for k=j

Sij is found by driving port j with an incident wave of voltage Vj+ , and

measuring the reected wave amplitude, Vi , coming out of port i. The incident waves on all ports except the jth port are set to zero, which means that all ports should be terminated in matched loads to avoid reections.
Sii is the reection coecient seen looking into port i when all other

ports are terminated in matched loads.

Sij , is the transmission coecient from port j to port i when all other

ports are terminated in matched loads.

University of Moratuwa

18 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Scattering Matrix

Example: Find the S parameters of the 3 dB attenuator circuit shown in

the gure.

Figure: A matched 3 dB attenuator with a 50 characteristics impedance

University of Moratuwa

19 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Scattering Matrix

We now show how the [S] matrix can be determined from the [Z] (or [Y ])

matrix, and vice versa. First, assume that the characteristic impedances, Z0n , of all the ports are identical. Then for convenience, we can set Z0n = 1. The total voltage and current at the nth port can be written as Vn In
Using the denition of [Z]
+ = Vn + Vn , + = In In .

(6) (7)

[Z][I] = [Z][V + ] [Z][V ] = [V ] = [V + ] + [V ], which can be written as ([Z] + [U ])[V ] = ([Z] [U ])[V + ], where [U ] is the unit or identity matrix.

University of Moratuwa

20 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Scattering Matrix

The scattering matrix can be given as

[S] = ([Z] + [U ])1 ([Z] [U ]).

For one-port network

(8)

S11 =
[Z] can be found in terms of [S]

z11 1 . z11 + 1

[Z] = ([U ] + [S])([U ] [S])1 .

University of Moratuwa

21 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Scattering Matrix

Reciprocal Networks and Lossless Networks

The impedance and admittance matrices are symmetric for reciprocal

networks, and purely imaginary for lossless networks.

Similarly, the scattering matrices for these types of networks have special

properties. By adding (6) and (7) we obtain

+ Vn

= =

[ +] V

1 (Vn + In ), 2 1 ([Z] + [U ])[I]. 2

By subtracting (6) and (7) we obtain

Vn

= =

[ ] V

1 (Vn In ), 2 1 ([Z] [U ])[I]. 2

EN4520-Microwave Communications Chandika Wavegedara

University of Moratuwa

22 / 35

Microwave Network Analysis

Scattering Matrix

It can be shown that

[V ] = ([Z] [U ])([Z] + [U ])1 [V + ], and [S] = ([Z] [U ])([Z] + [U ])1 . { }t ([Z] + [U ])1 ([Z] [U ])t . (9)

Taking the transpose of (9) gives

[S]t =

Now [U ] is diagonal, so [U ]t = [U ], and if the network is reciprocal, [Z] is

symmetric so that [Z]t = [Z]. The above then reduces to [S]t = ([Z] + [U ])1 ([Z] [U ]), which is equivalent to (8). We have thus shown that [S] = [S]t , for reciprocal networks.

University of Moratuwa

23 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Scattering Matrix

If the network is lossless, then no real power can be delivered to the

network.
If the characteristic impedances of all the ports are identical and assumed

to be unity, the average power delivered to the network is Pav = =

1 + t + 2 [V ] [V ]

1 Re{[V ]t [I] } 2 1 +t + 1 t [V ] [V ] [V ] [V ] = 0. 2 2

represents the total incident power while 1 [V ]t [V ] 2 represents the total reected power. For lossless junction we have 1 +t + 1 [V ] [V ] = [V ]t [V ] . 2 2

Using [V ] = [S][V + ]

[V + ]t [V + ] = [V + ]t [S + ]t [S] [V + ] .
For nonzero [V + ], [S + ]t [S] = [U ], or [S] = {[S]t }1 , [S] is a unitary

matrix.
University of Moratuwa
24 / 35 EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Scattering Matrix

We can express [S + ]t [S] = [U ] in summation form as

N k=1 Ski Skj = ij

for all i, j,

where ij = 1 if i = j and ij = 0 if i = j is the Kronecker delta symbol. Thus, if i = j N Ski Ski = 1,

k=1

while if i = j

N k=1

Ski Skj = 0.

The dot product of any column of [S] with the conjugate of that column

gives unity, while the dot product of any column with the conjugate of a dierent column gives zero (orthogonal).
If the network is reciprocal, then [S] is symmetric, and the same

statements can be made about the rows of the scattering matrix.

University of Moratuwa
25 / 35 EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Scattering Matrix

Example: A two-port network is known to have the following scattering

matrix [S] =

0.150o 0.8545o

0.85 45o 0.20o

] .

Determine if the network is reciprocal, and lossless. If port two is terminated with a matched load, what is the return loss seen at port 1? If port two is terminated with a short circuit, what is the return loss seen at port 1?

University of Moratuwa

26 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Scattering Matrix

Generalized Scattering Parameters

So far we have considered the scattering parameters for networks with the

same characteristic impedance for all ports.

This is the case in many practical situations, where the characteristic

impedance is often 50 . In other cases, however, the characteristic impedances of a multiport network may be dierent, which requires a generalization of the scattering parameters as dened up to this point. Consider the N -port network shown in the gure, where Z0n is the (real) + characteristic impedance of the nth port, and Vn and Vn , respectively, represent the incident and reected voltage waves at port n. In order to obtain physically meaningful power relations in terms of wave amplitudes, we must dene a new set of wave amplitudes as + an = Vn / Z0n , (10) bn = Vn / Z0n , (11) where an represents an incident wave at the nth port, and bn represents a reected wave from that port.
University of Moratuwa
27 / 35 EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Scattering Matrix

Figure: An N -port network with dierent characteristic impedances

University of Moratuwa

28 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Scattering Matrix

Then we have that

Vn In

+ = Vn + Vn = Z0n (an + bn ), 1 1 (an bn ). = (V + Vn ) = Z0n n Z0n

Now the average power delivered to the nth port is

Pn =

1 1 1 Re{Vn In } = |an |2 |bn |2 . 2 2 2

This is physically satisfying results, since it says that the average power

delivered through port n is equal to the power in the incident wave minus the power in the reected wave. + If expressed in terms of Vn and Vn , the corresponding result would be dependent on the characteristic impedance of the nth port.

University of Moratuwa

29 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

Scattering Matrix

A generalized scattering matrix can then be used to relate the incident

and reected waves [b] = [S][a], where the i, jth element of the scattering matrix is given by Sij = bi aj for k = j,
ak =0

and is analogous to the result for networks with identical characteristic impedance at all ports. Using (10) and (11) Vi Z0j Sij = Vj+ Z0i for k = j,
+ Vk =0

which shows how the S parameters of a network with equal characteristic impedance can be converted to a network connected to transmission lines with unequal characteristic impedances.
University of Moratuwa
30 / 35 EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

The Transmission Matrix

The Transmission (ABCD) Matrix

The Z, Y , and S parameter representations can be used to characterize a

microwave network with an arbitrary number of ports.

In practice many microwave networks consist of a cascade connection of

two or more two-port networks.

In this case it is convenient to dene a 2 2 transmission, or ABCD

matrix, for each two-port network.

The ABCD matrix of the cascade connection of two or more two-port

networks can be easily found by multiplying the ABCD matrices of the individual two-ports. The ABCD matrix is dened for a two-port network in terms of the total voltages and currents as shown in the gure and the following: V1 I1 or in matrix form as [ V1 I1 ] = = = [ AV2 + BI2 , CV2 + DI2 , A C B D ][ V2 I2 ] .

University of Moratuwa

31 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

The Transmission Matrix

Figure: (a) A two-port network; (b) a cascade connection of two-port networks.

Note that a change in the sign convention of I2 has been made from our

previous denitions, which had I2 as the current owing into port 2. The convention that I2 ows out of port 2 will be used when dealing with ABCD matrices so that in a cascade network I2 will be the same current that ows into the adjacent network shown in Figure (b).
University of Moratuwa
32 / 35 EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

The Transmission Matrix

In the cascade connection of two two-port networks shown in Figure (b),

we have that

[ [

V1 I1 V2 I2

] = ] =

[ [

A1 C1 A2 C2 ][

B1 D1 B2 D2

][ ][

V2 I2 V3 I3

] , ] . (13) (12)

Substituting (13) into (12) gives

V1 I1

] =

A1 C1

B1 D1

A2 C2

B2 D2

][

V3 I3

] ,

which shows that the ABCD matrix of the cascade connection of the two networks is equal to the product of the ABCD matrices representing the individual two-ports. A library of ABCD matrices for elementary two-port networks can be built up, and applied in building-block fashion to more complicated microwave networks that consist of cascades of these simpler two-ports.
University of Moratuwa
33 / 35 EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

The Transmission Matrix

University of Moratuwa

34 / 35

EN4520-Microwave Communications Chandika Wavegedara

Microwave Network Analysis

The Transmission Matrix

Relation to Impedance Matrix

Knowing the Z parameters of a network, one can determine the ABCD

parameters.
Thus, for a two-port network with I2 to be consistent with the sign

convention used with ABCD parameters V1 V2 we have that A B C = = = V1 V2 V1 I2 I1 V2 = Z11 /Z21 ,

I2 =0

= =

I1 Z11 I2 Z12 , I1 Z21 I2 Z22 ,

=
V2 =0

Z11 Z22 Z12 Z21 , Z21 I1 I2 = Z22 /Z21 .

V2 =0

= 1/Z21 , and D =
I2 =0

If the network is reciprocal (then Z12 = Z21 ), we can show that

AD BC = 1.

University of Moratuwa

35 / 35

EN4520-Microwave Communications Chandika Wavegedara