UTD Technical Reference Manual 페이지 1 / 23

SuperNEC
UTD Technical Reference Manual

Version 2.7

Document Status: Release

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 2 / 23

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 3 / 23

A1.Table of Contents
1. Introduction

2. Computations specific to plates

3. Computations specific to cylinders

3.1. Radius of curvature in the principal planes
3.2. Radius of curvature in plane of incidence
3.3. Finding the length of a geodesic path

4. Computations specific to diffraction

4.1. The diffraction transition function

5. UTD diffraction off a wedge

5.1. The UTD equation
5.1.1. The UTD edge diffraction coefficients
5.1.2. The a± function
5.1.3. The 3D distance parameters
5.1.4. The spreading factor
5.1.5. Computing the diffracted wavefront radii

6. UTD slope diffraction off a wedge

6.1. The UTD equation
6.1.1. The UTD slope diffraction coefficients
6.1.2. Common functions

7. UTD corner diffraction

7.1. The UTD equation
7.1.1. The UTD corner diffraction coefficients
7.1.2. The spreading factor
7.1.3. The corner distance parameter for spherical incidence

8. UTD reflection off a cylinder

8.1. The UTD equation
8.1.1. The UTD reflection coefficients
8.1.2. The 3-D distance parameters
8.1.3. The 3D spreading factor
8.1.4. Computing the reflected wavefront radii

9. UTD diffraction off a cylinder

9.1. The UTD equation
9.1.1. The UTD diffraction coefficients
9.1.2. The conservation of energy term
9.1.3. The 3-D distance parameters
9.1.4. The 3-D spreading factor
9.1.5. Computing the diffracted wavefront radii
9.1.6. Evaluating the transition function
9.2. The UTD equation for grazing incidence
9.2.1. The UTD diffraction coefficients for grazing
9.2.2. Computing the grazed wavefront radii

10. References

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 4 / 23

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 5 / 23

1. Introduction
SuperNEC (SNEC) is an object-oriented version of the FORTRAN program NEC-2. This technical manual is therefore
very similar to the NEC-2 manual. In fact, many sections have been copied verbatim from the NEC-2 manual. The
reason this document has been produced (as opposed to referring the reader to any readily available NEC-2 manual) is
because new theory will be added to SNEC and some features that exist in NEC-2 will not be implemented in SNEC.
The easiest way of keeping an up to date reference of the theory implemented in SNEC is to start with the an electronic
version of the NEC-2 manual and modify the document as the SNEC code evolves. Many extensions have been made to
SNEC in prototype form. These features include hybridisation with UTD, fast iterative solvers, MBPE amongst other
features. When these extensions are formally incorporated into the SNEC program, then the theory behind the extension
will be added to this manual.

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 6 / 23

2. Computations specific to cylinders

This sections details some of the computations that apply to cylinders.

2.1. Radius of curvature in the principal planes

The principal planes of the cylinder were chosen such that U$ 1 is tangent to the cylinder in the XY plane and U$ 2 is in
the direction of the z-axis of the cylinder. The radius of curvature in the XY plane for an elliptical cylinder is given by
3
(1)
(a22 cos2 v + a12 sin 2 v) 2
a1 =
a1a 2

2.2. Radius of curvature in plane of incidence

( i
)
The plane of incidence is defined as the plane containing s$ , n$ .In general the radius of curvature in a plane at an angle
a from the principal surface axes is given by Euler’s theorem as :

1 sin 2 a cos2 a (2)

= +
at a1 a2

For the cylinder, a1 is as given in section 2.1, whilst a 2 = ¥ . The angle a is computed as :

t1 = - s$ i · U$ 1 (3)
t 2 = - s$ i · U$2

æ t1 ö
a = tan -1 ç ÷
è t2ø

2.3. Finding the length of a geodesic path

This details how to find the distance along the surface of the cylinder from elliptical angle v1 to v2 in a given
direction. The arc length is given by :
v2
(4)
ò (a12 sin 2 a + a 22 × cos2 a ) da
v1
arcLength =
sin q
where

q is the elevation angle of the path traversed between v1 and v2 .

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 7 / 23

3. Computations specific to diffraction

3.1. The diffraction transition function
The transition function used in both the UTD edge and curved surface diffraction is defined by [ 1]:
¥
2 (5)
F ( x ) = 2 j x × e jx ò e - ju du
x
For large argument

j 3 j15 (6)
F ( x) = 1 + - 2 - 3
2x 4x 8x
For small argument

F ( x) = [ ]
px - 2 xe jp / 4 - (2 / 3) x 2 e - jp / 4 e jp / 4 e jx (7)
For argument . , the interpolation scheme used for numerical computation is :
0.3 £ x £ 55

F ( x ) = F ( xi ) +
[ F ( xi +1 ) - F ( xi )] ( x - x ) (8)
i
( xi +1 - xi )
where the values for the iterative scheme are obtained from :

xi F ( xi ) [ F ( x i +1 ) - F ( x i ) ] / ( x i +1 - x i )
Real Imaginary Real Imaginary
0.3 0.0 0.0 0.5729 0.2677
0.5 0.5195 0.0025 0.6768 0.2682
0.7 0.3355 -0.0665 0.7439 0.2549
1.0 0.2187 -0.0757 -0.8095 0.2322
1.5 0.1270 -0.680 0.8730 0.1982
2.3 0.0638 -0.0506 0.9240 0.1577
4.0 0.0246 -0.0296 0.9658 0.1073
5.5 0.0093 -0.0163 0.9797 0.0828

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 8 / 23

4. UTD diffraction off a wedge

4.1. The UTD equation
The field diffracted from a wedge is given by :
r
E d ( p) = E i (Qe ) · D × A( s) × e - jks (9)
where

E i (Qe ) is the field incident on the edge at point Qe

D is the dyadic diffraction coefficient.
A(s) is the spreading factor.
Written in matrix form in the edge coordinate system :

é E bd ù é- Ds 0 ù é E bi (Qe ) ù r (10)
ê dú= ê ú ê i ú A( s) × e - jks
êë E f úû ë 0 - Dh û êë E f (Qe ) úû
where

E bd is the diffracted field in the b direction (parallel to the plane of diffraction)

E fd is the diffracted field in the f direction (perpendicular to the plane of diffraction)
Ds is the soft / parallel diffraction coefficient
Dh is the hard / perpendicular diffraction coefficient
E bi (Qe ) is the incident field in the b direction (parallel to the plane of incidence)
E fi (Qe ) is the incident field in the f direction (perpendicular to the plane of incidence)

4.1.1. The UTD edge diffraction coefficients

The soft and hard diffraction coefficients are given by :

Ds,h ( Li , Lro , Lrn , f r , f i , b i , n) = D1 + D2 + Rs,h ( D3 + D4 ) (11)

for all angles other than grazing incidence where

Ds ( Li , Lro , Lrn , f r , f i , b i , n) = 0 (12)

i ro rn r i i
Dh ( L , L , L , f , f , b , n) = D1 + D2
Where :

Li , Lro , Lrn are the distance parameters for the incident shadow boundary and the reflection shadow boundary
for the o-face and the n-face respectively.
f r is the diffraction angle with respect to the o-face 0 £ f r £ np
f i is the incident angle with respect to the o-face 0 £ f i £ np
b i is the angle between the incident ray and the edge
n is the wedge angle number.
Rs,h = m1 are the soft and hard reflection coefficients of the surfaces of the wedge at the edge.
D1,2 ,3,4 are the components of the diffraction coefficients.
The diffraction components for 3-D diffraction are given by :

(13)

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 9 / 23

- e - jp / 4 é p + (f r - f i ) ù
D1 =
2n 2 pk sin b i cot
ê
êë 2n úû
[
ú F kLi a + (f r - f i ) ]

- e - jp / 4 (
é p - f r - fi ) ùú F kL a (14)
D2 =
2n 2 pk sin b i
cot ê
êë 2n úû
[ i -
(f r
- fi )]

- e - jp / 4 (
é p + f r + fi ) ùú F kL (15)
D3 =
2n 2 pk sin b i
cot ê
êë 2n úû
[ rn
(
a + f r + fi )]

- e - jp / 4 (
é p - f r + fi ) ùú F kL (16)
D4 =
2n 2 pk sin b i
cot ê
êë 2n úû
[ ro
(
a - f r + fi )]
where

F [ X ] is the transition function.

k is the wave number 2p / l
a ± ( f r ± f i ) is defined in section 4.1.2

4.1.2. The a function

The a ± function used in the diffraction coefficient calculation is defined as :

æ 2npN ± - Df ± ö (17)
a ± ( Df ± ) = 2 cos2 ç ÷
è 2 ø
where

Df ± = f r ± f i (18)
and N ± are the integers that most nearly satisfy the equations :

2 pnN + - (f r ± f i ) = p (19)

2 pnN - - (f r ± f i ) = - p
The values of N + , N - are computed as :

æ p + (f r ± f i ) ö (20)
N = RoundUpçç
+ ÷
÷
è 2 p n ø
æ - p + (f r ± f i ) ö
N = RoundUpçç
- ÷
÷
è 2 p n ø
NOTE : ( a + , N + ) are associated with the n-face and ( a - , N - ) are associated with the o-face.

4.1.3. The 3D distance parameters

i ro rn
The distance parameters L , L , L are defined here [ 3]:

(21)

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 10 / 23

i s(rie + s)r1i ri2

L = sin 2 b
rie (r1i + s)(ri2 + s)
where

r1i is the radius of curvature of the incident wavefront at Qe in the plane of incidence i.e., the plane containing
( s$ i , n$) .
ri2 is the radius of curvature of the incident wavefront at Qe transverse to the in the plane of incidence.
rie is the radius of curvature of the incident wavefront at Qe in the edge-fixed plane of incidence i.e., the
i
plane containing ( s$ , e$) .

s(r ro
e
.n
+ s)r1ro.n r ro
2
,n
(22)
Lro ,n = sin 2 b
r ro ,n
e ( r1
ro ,n
+ s)(r ro
2
,n
+ s)
where

r1ro,2,n are the principal radii of curvature of the reflected wavefront at Qe from the o-face and n-face
respectively.
r ro
e
,n
are the radii of curvature of the reflected wavefront at Qe in the plane containing the reflected ray and
r
the edge i.e., the plane containing ( s$ , e$ ) .

For an edge formed by the junction of two flat surfaces, the radii of curvature of the reflected wavefront are equal to the
radii of curvature of the incident wavefronts. Thus

r1r = r1i (23)

r r2 = ri2
r re = rie

4.1.4. The spreading factor

The spreading factor for the edge diffracted field is given by :

r (24)
A( s) =
s( s + r)
where
r is the edge caustic distance.
i
For flat plates the edge caustic is equal to r e , i.e., the radius of curvature of the incident wavefront at Qe in
i
the plane containing ( s$ , e$) .

4.1.5. Computing the diffracted wavefront radii

The principal directions of the diffracted wavefront are :

1.
$
The plane containing the edge and the reflected ray ( X 2 )
r

The plane transverse to plane 1 ( X$ 1 ).

r
2.

The two unit vectors are computed as :

X 1r = n$ × cos f r - t$ × sin f r (25)

X r = f$ ´ s$ r
2 d

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 11 / 23

where
n$ is the unit normal of the o-face
t$ is the surface tangent of the o-face
f r is the angle of the reflected plane ( s$ r , e$) measured from the o-face.
The radii of curvature of the reflected wavefront at the point of observation are :

r1r = s r (26)

r r2 = s r + rie

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 12 / 23

5. UTD slope diffraction off a wedge

This section details the aspects of slope diffraction that differ from edge diffraction.

5.1. The UTD equation

The slope diffracted field from a wedge is given by :

sd 1 ¶Ds,h ¶E i (Qe ) - jksr (27)

E ( p) = × × × A( s ) × e
jks i sin b ¶f i ¶n$
where

E i (Qe ) is the field incident on the edge at point Qe

n$ is the normal to the edge
A(s) is the spreading factor.
Written in matrix form in the edge coordinate system :

é ¶E bi (Qe ) ù (28)
é E bd ù é- Dssd 0 ù ê ¶n$ ú r
ê dú= ê ê i ú × A( s) × e - jks
sd ú
êë E f úû ë 0 - Dh û ê ¶E f (Qe ) ú
ê ú
ë ¶n$ û
where

E bd is the slope diffracted field in the b direction (parallel to the plane of diffraction)
E fd is the slope diffracted field in the f direction (perpendicular to the plane of diffraction)
Dssd is the soft / parallel slope diffraction coefficient
Dhsd is the hard / perpendicular slope diffraction coefficient
¶E bi (Qe )
¶n$ is the derivative of the incident field in the b direction (parallel to the plane of incidence) with
respect to n$ .

¶E fi (Qe )
¶n$ is the derivative of the incident field in the f direction (parallel to the plane of incidence) with
respect to n$ .

5.1.1. The UTD slope diffraction coefficients

The soft and hard diffraction coefficients are given by :

¶E i (Qe ) (29)
Dssd,h ( Li , Lro , Lrn , f r , f i , b i , n) = D1sd + D2sd - Rs,h ( D3sd + D4sd )
¶n$
for all angles other than grazing incidence where

¶E i (Qe ) (30)
Dssd ( Li , Lro , Lrn , f r , f i , b i , n) = D1sd + D2sd
¶n$
i ro rn r ¶E i (Qe )
i i
Dh ( L , L , L , f , f , b , n) =0
¶n$
Where :

Li , Lro , Lrn are the distance parameters for the incident shadow boundary and the reflection shadow boundary
for the o-face and the n-face respectively.

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 13 / 23

f r is the diffraction angle with respect to the o-face 0 £ f r £ np

f i is the incident angle with respect to the o-face 0 £ f i £ np
b i is the angle between the incident ray and the edge
n is the wedge angle number.
Rs,h = m1 are the soft and hard reflection coefficients of the surfaces of the wedge at the edge.
D1sd,2 ,3,4 are the components of the slope diffraction coefficients.
The slope diffraction components for 3-D diffraction are given by :

- e - jp / 4 (
é p + f r - fi ) ùú F ¶E i (Qe ) (31)
D1sd = 2
4n 2 pk sin b i
2
cos ec ê
êë 2n úû
s [kL a (f
i + r
-f i
)] ¶n$ni

e - jp / 4 (
é p - f r - fi ) ùú F ¶E i (Qe ) (32)
D2sd = 2
4n 2 pk sin b i
2
cos ec ê
êë 2n úû
s [kL a (f
i - r
-f i
)] ¶n$oi

- e - jp / 4 (
é p + f r + fi ) ùú F ¶E i (Qe ) (33)
D3sd = 2
4n 2 pk sin b i
2
cos ec ê
êë 2n úû
s [kL a (f
rn + r
+f i
)] ¶n$nr

e - jp / 4 (
é p - f r + fi ) ùú F ¶E i (Qe ) (34)
D4sd = 2
4n 2 pk sin b i
2
cos ec ê
êë 2n úû
s [ ro
kL a f + f-
( r i
)] ¶n$or
where

Fs [ X ] = 2 jX [1 - F [ X ]] is given in terms of the transition function F [ X ] .

k is the wave number 2p / l
a ± ( f r ± f i ) is defined in section 4.1.2

5.1.2. Common functions

See the appropriate subsections of section 4 for

· The a ± function

· Computing the 3D distance parameters.

· Computing the 3D spreading factor.
· Computing the diffracted wavefront radii.

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 14 / 23

6. UTD corner diffraction

The corner diffracted wave for spherically incident plane waves is described in this section.

6.1. The UTD equation

The UTD equation for corner diffracted fields is given by :

sin b c sin b 0c (35)

E bc,f = - E bi ,f (Qc ) × Cs,h (Qc ) ×
cos b 0c - cos b c
[ ]
× F kLc a ( p + b 0c - b c ) ×

l - jp / 4 r
×e × A( s) × e - jks
2p
where

E bc,f are the b and f components of the corner diffracted field.

E bi ,f (Qc ) are the b and f components of the field incident at Qc .
The angles b c and b 0c are illustrated in Figure 1.
Lc is the distance parameter associated with the corner.
F [ X ] is the transition function.
Cs,h (Qe ) are the corner diffraction coefficients.
A( s) is the spreading factor.
æ b - b 0c ö
a( p + b 0c - b c ) = 2 cos2 ç c ÷
è 2 ø
Error! Filename not specified.
Figure 1: The Geometry for corner diffracted fields [ 1].

6.1.1. The UTD corner diffraction coefficients

The soft and hard diffraction coefficients are given by :

(
Cs,h (Qe ) = D1 × F [ X 1 ] + D2 × F [ X 2 ] + Rs,h D3 × F [ X 3 ] + D4 × F [ X 4 ] ) (36)
or for grazing incidence

Cs (Qe ) = 0 (37)
[ ]
Ch (Qe ) = D1 × F X 1 + D2 × F X 2 [ ]
where
Rs,h = m1 are the soft and hard reflection coefficients of the surfaces of the wedge.
D1,2 ,3,4 are the components of the diffraction coefficients as defined in equations.(13) to (16).
and

(38)

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 15 / 23

é Li a ( f r - f i ) / l ù
[ ]
F X1 = F ê ú
ë kLc a ( p + b 0c - b c ) û
é Li a ( f r - f i ) / l ù
[ ]
F X2 = Fê ú
ë kLc a ( p + b 0c - b c ) û
é Lrn a ( f r + f i ) / l ù
[ ]
F X3 = Fê ú
ë kLc a ( p + b 0c - b c ) û
é Lro a ( f r + f i ) / l ù
[ ]
F X4 = Fê ú
ë kLc a ( p + b 0c - b c ) û

6.1.2. The spreading factor

The spreading factor for the corner diffracted field is given by :

1 (39)
A( s) =
sr

6.1.3. The corner distance parameter for spherical incidence

The distance parameter Lc is given by :

s ic s rc (40)
Lc =
s ic + s rc
where

s ic is the distance from the source to the corner

s rc is the distance from the corner to the observation point

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 16 / 23

7. UTD reflection off a cylinder

The SNEC MOM/ITD hybrid program uses the UTD theory for computing the electric field reflected off the curved
surface of a cylinder. The implemented equations are outlined in this section.

7.1. The UTD equation

The field reflected off a cylinder from an electric field source is given by [ 1] as
r
E r ( p) = E i (Qr ) × Rs,h × A( s) × e - jks (41)
where

E i (Qr ) is the E-field incident on the cylinder at Qr

Rs ,h are the soft and hard UTD reflection coefficients
A( s) is the 3-D spreading factor.

7.1.1. The UTD reflection coefficients

The UTD soft and hard reflection coefficients [ 2] are given by :

- 4 - jp / 4 - jx 3 /12 é - F ( X p ) ì p * ( x) üù (42)
Rs.h = - e e ê + í * ýú
xp êë 2x p î q ( x) þúû
where

p * ( x) and q * ( x) are the soft and hard Fock scattering functions.

F ( X p ) is the diffraction transition function.
X p = 2 kL p cos2 q i is the transition function argument.
x = -2m(Qr ) cos q i
and

cos q i = n$ · s$ r where n$ is the unit normal at the point of incidence and s$ r is the reflected ray.
2p
k=
l is the wave number
1
æ ka (Q ) ö 3
m(Qr ) = ç 0 r ÷
è 3 ø is the curvature parameter with a 0 (Qr ) the radius of curvature of the surface in the
i
plane of incidence ( n$ , s$ ) .
L p is the 3-D distance parameter.
The reflection coefficients are computed in the following form :

- 2 - jp / 4 - jx 3 /12 é - F ( X p ) ì p * ( x ) üù (43)
Rs.h =- e e ê + 2 p í * ýú
xp êë x 2 î q ( x) þúû

7.1.2. The 3-D distance parameters

The distance parameter [ 2] for the computation of the transition function argument is given by :

r1i ri2 s r (r r2 + s r ) (44)

Lp =
(r1i +s r
)(r1i +s )r
r r2
where

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 17 / 23

r1i is the radius of curvature of the incident wave in the plane of incidence.
ri2 is the radius of curvature of the incident wave transverse to the plane of incidence.
r1r is the radius of curvature of the reflected wave in the plane of reflection.
r r2 is the radius of curvature of the reflection wave transverse to the plane of reflection.
s r = P - Q r the distance from point of incidence to the observation point.

7.1.3. The 3D spreading factor

The spreading factor for the reflected wave is given by :

r1r r r2 (45)
A( s) =
(r1r + s r )(r r2 + s r )
where

r1r is the radius of curvature of the reflected wave in the plane of reflection.
r r2 is the radius of curvature of the reflection wave transverse to the plane of reflection.
s r = P - Q r the distance from point of incidence to the observation point.

7.1.4. Computing the reflected wavefront radii

r r
The principal radii of curvature of the reflected wavefront r1 , r 2 and the principal directions (axes) of the wavefront
are given in this section. The plane of incidence may be different from the principal planes of the reflecting surface, so
r r
that the principal directions of the incident wavefront are quite distinct from those of the reflecting surface. r1 , r 2 are
given as [ 3] :

1 1é 1 1ù 1 (46)
= ê + ú+
r1r 2 ë r1i ri2 û f 1
and

1 1é 1 1ù 1 (47)
r = ê i + i ú+
r 2 2 ë r1 r 2 û f 2
where
2
1 cos q i é G22 + G122 G21
2
+ G112 ù (48)
= 2 ê + ú±
f 1,2 G ë a1 a2 û
2
1 éæ 1 1ö
êç i - i ÷ +
2 êè r1 r 2 ø
ë
2
æ 1 1 ö 4 cos q i é G22 - G122 G21
2
- G112 ù
ç i - i÷ ê + ú+
è r1 r 2 ø 4 G 2 ë a1 a2 û
2 2
4 cos2 q i ìïæ G22 4 G üï ù
2 2
+ G122 G21
2
+ G112 ö
íç + ÷ - ý
G
4
ïîè a1 a2 ø a1a 2 ï úû
þ
The matrix G is defined in terms of the principal planes of the incident wavefront ( X$ 1i , X$ 2i ) and the principal
directions of the curved surface (U$ 1 )
, U$ 2 .The principal directions of the cylinder have been chosen such that : U$ is
1

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 18 / 23

tangent to the cylinder in the XY plane and U$ 2 is in the direction of the z-axis of the cylinder. Note : All computations
are done in the cylinder coordinate system.

é X$ 1i · U$ 1 X$ 1i · U$ 2 ù (49)
G = ê $i $ ú
ë X 2 · U1 X$ 2i · U$ 2 û
The determinant of G is represented as G .

( r r)
The principal directions of the reflected wavefront are given in terms of the vectors x$1 , x$ 2 , which are unit vectors
$i $i
perpendicular to the reflected ray. The are determined by reflecting the unit vectors ( X 1 , X 2 ) in the plane tangent to
the surface at Qr i.e.,

( )
x$1r,2 = X$ 1i,2 - 2 n$ · X$ 1i,2 n$ (50)
The principal directions of the reflected wavefront are then given by :

éæ r 1ö ù (51)
êç Q22 - r ÷ x$1r - Q12
r r
x$ 2 ú
êëè r1 ø úû
X$ 1r =
2
æ r 1ö r 2
ç Q22 - r ÷ + (Q12 )
è r1 ø
and

X$ 2r = - s$ r ´ X$ 1r (52)
Where
2 2
r 1 2 cos q i é G22 G21 ù (53)
Q11 = i + 2 ê + ú
r1 G ë a1 a2 û

r r - 2 cosq i é G22 G12 G11G21 ù (54)

Q12 = Q21 = 2
ê + ú
G êë a1 a 2 úû

2 2
r 1 2 cos q i é G12 G11 ù (55)
Q22 = i + 2 ê + ú
r2 G ë a1 a2 û

NOTE : The principal directions of the wavefront are distinct from the principal directions associated with the

reflected wave directions i.e., X$ 1r ¹ e$ parallel

r $r r
and X 2 ¹ e$ perpendicular

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 19 / 23

8. UTD diffraction off a cylinder

The SNEC MOM/UTD hybrid program uses the UTD theory for computing the electric field diffracted from the curved
surface of a cylinder. The implemented equations are outlined in this section.

8.1. The UTD equation

The field diffracted off a cylinder from an electric field source is given by [ 1] as

dh(Q1 ) r (56)
E d ( p) = E i (Q1 ) × Ts,h × × A( s) × e - jks
dh(Q2 )
where

E i (Q1 ) is the E-field incident on the cylinder at Q1 .

Ts,h are the soft and hard UTD diffraction coefficients.
The square root term is the conservation of energy term as wave creeps around cylinder.
A( s) is the 3-D spreading factor.

8.1.1. The UTD diffraction coefficients

The UTD soft and hard diffraction coefficients [ 2] are given by :

2 - jp / 4 - jkt é - F ( X d ) ì p * ( x) üù (57)
Ts.h = - m(Q1 )m(Q2 ) × ×e ×e ê + í * ýú
k êë 2x p î q ( x) þúû
where

p * ( x) and q * ( x) are the soft and hard Fock scattering functions.

F ( X d ) is the diffraction transition function.
kLd x 2
Xd =
2m(Q1 )m(Q2 ) is the transition function argument.
t ( Q2 )
m( t) dr
x= ò a ( t )
dt dt =
dg
dg
t ( Q1 ) where

r ( g ) = - a1 sin g × x$ + a 2 cos g × y$
and

cos q i = n$ · s$ r where n$ is the unit normal at the point of incidence and s$ r is the reflected ray.
2p
k=
l is the wave number
1
æ ka (Q ) ö 3
m(Qr ) = ç 0 r ÷
è 3 ø is the curvature parameter with a 0 (Qr ) the radius of curvature of the surface in the
i
plane of incidence ( n$ , s$ ) .
Ld is the 3-D distance parameter.
The reflection coefficients are computed in the following form :

1 - jp / 4
é
- jkt - F ( X d )
ì p * ( x ) üù (58)
Ts.h = - m(Q1 )m(Q2 ) × ×e ×e ê + 2 p í * ýú
pk êë x 2 î q ( x) þúû
Where the argument for the Fock scattering functions is computed as :

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 20 / 23

r ( g ) = - a1 sin g × x$ + a 2 cos g × y$
t ( Q2 )
m( t) dr
x= ò a ( t )
dt dt =
d g
dg = a12 sin 2 g + a 22 cos2 g × dg
t ( Q1 ) where

Thus
1
2 2 g
æ k sin q ö 3 1
x=ç ÷ ( a1a 2 ) 3 × ò dg
è 2 ø a 2
sin 2
g + a 2
cos 2
g
g1 1 2

a( t) is the radius of curvature is the direction of incidence.

8.1.2. The conservation of energy term

The conservation of the energy flux in the surface-ray strip from Q1 to Q2 is given by :

dh(Q1 ) (59)
EC =
dh(Q2 )
This is computed (for spherically incident waves) as :

dh(Q1 ) ri2 (60)

EC = =
dh(Q2 ) ri2 +t
where t is the creep distance along the surface geodesic between Q1 and Q2 .

8.1.3. The 3-D distance parameters

The distance parameter [ 2] for the computation of the transition function argument is given by :

r1i ri2 s r (r r2 + s r ) (61)

Ld =
(r1i +s r
)(r1i r
+s ) r r2
where

r1i is the radius of curvature of the incident wave in the plane of incidence.
ri2 is the radius of curvature of the incident wave transverse to the plane of incidence.
r1r is the radius of curvature of the reflected wave in the plane of reflection.
r r2 is the radius of curvature of the reflection wave transverse to the plane of reflection.
s r = P - Q 2 the distance from point of incidence to the observation point.

8.1.4. The 3-D spreading factor

The 3D spreading factor for diffraction is given by :

r r2 (62)
A( s) =
s r (r r2 + s r )
where

r r2 is the radius of curvature of the reflection wave transverse to the plane of reflection.
s r = P - Q 2 the distance from point of incidence to the observation point.

8.1.5. Computing the diffracted wavefront radii

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 21 / 23

The two principal radii of curvature of the diffracted wave front at the point of observation is computed as follows :

r1r = rib + s r + t (63)

where t is the distance the wave crept around the cylinder and

1 sin 2 a cos2 a (64)

= +
rib r1i ri2
$
i.e., r b is the radius of curvature of the incident wavefront in the plane tangent to the surface. Plane containing ( t$ , b ) .
i

t$ is the vector tangent to the surface in the plane of incidence, b$ is the vector perpendicular to t$ on the surface and n$
is the normal to the surface. All vectors are computed at Q2 .

r r2 = s r (65)

The principal directions X 1r and X 2r are computed as :

X 1r = b$ = s$ r ´ n$Q2 (66)
and

X 2r = n$Q2 (67)

8.1.6. Evaluating the transition function

See section 3.1.

8.2. The UTD equation for grazing incidence

The field diffracted off a cylinder at grazing incidence from an electric field source is given by the sum of half the
incident field and the diffracted field modified for angles of grazing incidence at the observation point.

é sign r1rd r rd r2
rg ù r
(68)
d i 2 g
E ( p) = E (Q1 ) ê + Ts , h × ú × e - jks
ê 2 (r1rd + s r )(r rd r
2 +s )
rg
s r (r 2 + s r ) úû
ë
where

E i (Q1 ) is the E-field incident on the cylinder at Q1 , the point of grazing incidence
g
Ts,h are the soft and hard UTD diffraction coefficients modified for the point of grazing
Sign is positive if n$ · s$ r > 0 and negative otherwise
r1rd,2 are the radii of curvature of the direct field at Q1 in the plane of grazing for 1 (tangential to surface) and

in the plane transverse to grazing for 2 (direction of normal to surface). It is noted that r1rd,2 = r1i ,2
rg
r1,2 are the radii of curvature of the diffracted field in the same planes as above. In this case r1rg = 0 if the
rg rd i
point of reference is taken to be Q1 and r 2 = r 2 = r 2 .
The first square root term is the 3-D spreading factor for the incident field
The second square root term is the spreading factor for the diffracted field

The field is computed as :

é sign r1i s r ù r r2 (69)

d i g - jksr
E ( p) = E (Q1 ) ê + Ts , h
ú × × e
êë 2 (r1i + s r ) r r r
úû s (r 2 + s )
where the term in square brackets is grouped to form a modified diffraction coefficient.

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 22 / 23

8.2.1. The UTD diffraction coefficients for grazing

The UTD soft and hard diffraction coefficients for grazing are given by :

g 2 - jp / 4 ì p * (0) ü (70)
Ts,h = m(Q1 ) × ×e ×í * ý
k î q (0) þ
where

p * (0) and q * (0) are the soft and hard Fock scattering functions.
1
æ ka (Q ) ö 3
m(Qr ) = ç 0 1 ÷
è 3 ø is the curvature parameter with a 0 (Qr ) the radius of curvature of the surface in the
i
plane of incidence ( n$ , s$ ) .

8.2.2. Computing the grazed wavefront radii

The grazing wavefront radii and planes are the same as the diffracted values, except the creeping distance is zero. See
section 8.1.4.

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22
UTD Technical Reference Manual 페이지 23 / 23

9. References
[ 1] McNamara D.A., Pistorius C.W.I, Malherbe J.A.G “Introduction to the uniform geometrical theory of
diffraction”, Artech House, 1990.
[ 2] Pathak P.H, Burnside W.D, Marhefka R.J, “A Uniform GTD analysis of the diffraction of electromagnetic
waves by a smooth convex surface”, OEEE Trans. Antennas Propagat., Vol. AP-28, pp. 631-642, Sept.
1980.
[ 3] Kouyoumjian R.G, Pathak P.H, “A uniform geometrical theory of diffraction for an edge in a perfectly
conducting surface”, Proc IEEE, Vol. 62, pp. 1448-1461, Nov. 1974

http://www.supernec.com/manuals/snutdtrm.htm 2009-04-22