Indian Journal of Radio & Space Ph ysics Vol. 28 , February 1999, pp. 43-52

Simulation of secondary radiation pattern of MSMR antenna

S B Sharma & C Sriharsha

Sensors Antenna and Feed Group , Space Appli cations Centre, ISRO, Ahmedabad 380053

Received 2 J a lll/QI)' 1998; revised received 11 September 1998: accepted 13 October 1998

Simulati on of scco ndary radiation pattern of the multi frequency scanning microwave radiometer (MSMR) has been presented in thi s paper. Simulated patterns are used, as many practi cal difficulties have been encountered during measurement s. Apert ure integration in conj uncti on with geometric theory of diffraction has been used for simulation of secondary pattern of onset parabolic refl ector. The simulated and measured results are presented and fairly good agreement has been observed between them.

1 Introduction The multi frequency scanning microwave

radiometer (MSMR) to be launched on board IRS-P4 satellite will find its applicati on in estimation and monitoring of a number of geophys ical parameters re lated to the land , ocean and atmosphere. The an tenna forms the most critical part of thi s radiometer. The MSMR antenna system is required to receive thermal mi crowave radi ation emi ss ions from earth at four frequency bands, viz. 6.6 GHz, 10.65 GHz. 18.0 G Hz and 2 1.0 GHz. These four frequenc ies are received in two ports each for measurIn g horizontall y and ve rtically polarized energy . A hi gh degree of cross-polar isolation is required at each port. A high beam e ffici ency is al so required for this antenna and the beam is required to scan about the yaw axi s (axi s of the feed) . In order to meet these requirements an offset parabolic reflector having a large focus-to-diameter ratio (j /d) with a common feed for all the eight ports and having scanning mec hanism about the feed axis has been designed .

A bri ghtness tempe rature accuracy of I K is required to estimate the vanous geophysical parameters. In order to get brightness temperature of thi s accuracy the radiation pattern has to be measured in a very small interval at the boresight at each frequency and polarization and for a number of scan angles.' However, such measurements at high frequency and fo r an antenna having a large aperture diameter in open test range are not possible, as it involves making measurements at a large distance in order to avoid phase errors associated with near-field measurements . No test range, in thi s country, has the

capability for making far fi e ld measurements at such high frequency and for such large d iameter antennas. Compact range measurements are al so not possible because of large size and we ight of thi s antenna and its support stmcture . Keeping up with the tre':1d of shifting the re liance on numerical methods from expen sive, time consummg and inaccurate measurements, it was decided to explore the possibility of s imulati on of secondary radiation pattern.

In the present work the secondary radiation pattern has been computed from measured primary radiation pattern using the aperture integration method for findin g the pattern at main lobe and first fe w s ide lobes, and geome.tric theory of diffraction has been used to find pattern at far sidelobes. The measured resu lts at 30 m range has been compared with the simulated pattern at 30 m distance and was found to have a good agreement up to 20 dB level.

2 Description of MSMR antenna configuration

The geometry of MSMR antenna is shown in Fig. I. The Nf

44 INDIAN J RADIO & SPACE PHYS, FEBRUARY 1999

MECHAlliCAL CENTER

I o o '"

f ~~ " ; ~

6~ L--~ _ _ _ _ _ _

o

1440 --------'

OFF SET REFLECTOR WITH FEED

Fig. I-Geometry of MSMR antenna

3 Computation of secondary radiation pattern A brief description of the analysis involved in

aperture integration method for simulating the secondary pattern near the main lobe and first few side lobes, and the technique used for finding the diffracted fields from the rim of the reflector using geometrical theory of diffraction are given as follows.

3.1 Aperture integration method

A parabol ic reflector antenna has the property that al l path lengths from the focal point to the reflector and on to the aperture plane are the same.

The copol ar primary radiation pattern of the feed is given by

Er=e xp( -jkor)lr{ e(Oo,$o)(aoo sin $o+aoo cos $o)}

(1)

where, e(eo,$o) is the mea'sured copolar primary pattern, r the distance from the phase centre to the observation point, ko the wave number, and eo, $0 are, respectively, the e levation and the azimuth angles of

the feed and a 90, a¢o are the unit vectors in the spherical co-ordinate system.

The reflected field from the reflector surface is given by

Er= -Er+ 2(n.Er)n ... (2)

where, n is the outward nonnal of the reflecting surface.

The electric field on the aperture of the reflector can thus be found follow ing the methods illustrated by Collinl.

Any small displacement of the phase centre from the focus will result in a small change in phase of aperture field due to small path length perturbation.

From the equivalence principle the magnetic surface currents at the aperture plane can be found and the far field copolal" radiation pattern can be computed .

3.2 Computation of secondary pattern using GTD

Aperture integrati on (AI) method fa il s to yield satisfactory results after main lobe and a first few sidelobes. Therefore, geometri c theory of diffraction (GTD) has been used to predict secondary pattern in the region where Al cannot be used . T he switch-over

from AI to GTD is done when e >11:j A W , where A W is the aperture width o f the reflector in terms of A, as at this value of e, the va lue of the field obtained by AI and GTD are found to be the same. The value of a field in a diffracted ray is determined from the incident field with the aiel of appropriate diffraction coefficients, that is, a dyadic for electromaglletic fields. The diffraction and attenuation coefficients are usually de termined from the asymptotic solutions of the simplest boundary value problems, which have the same local geometry at the points of diffraction as the object at the point of interest The primary objective of using GTD is to resolve each problem into smaller components, each representing a canonical geometry of a known solution .

The GTD analysis of the reflector is si milar to that of diffraction by a nat plate in which each segment is treated as an edge of a flat plate which is tangential to reflector surface. Uniform GTD techniquesV are used to calculate the edge-d iffracted and slope-diffracted field s. The reflector rim is treated as. a piecewise linear segments.

The slope-diffracted fields are calculated in a similar way except that the slope-diffraction

coefficients oDs 10$' and oDh /0$' and the slope-diffraction coefficients oE,lo Il of the incident field at the edge are used.

Since each rim segment is small, the diffractions from its two end-points are sign ificant. These diffractions a re calculated by using the comer diffraction analysis. T he comer diffraction

;

SHARMA & SRIHARSHA: SIMULATION OF RADIATION PATTERN OF MSMR ANTENNA 45

compensates for the discontinuity which occurs when the diffraction point moves off to the rim segment.

4 Numerical results and conclusions The various input parameters given to simulate the

secondary pattern like focus, diameter of the reflector, etc . were as per the physical dimension of the antenna system. The primary radiation pattern was measured at various

46

a: w 3:

INDIAN J RADIO & SPACE PHYS, FEBRUARY 1999

o~------------------------~~--------------------------

-- SIMULATED

- - - MEASURED -10

~ -30

-m 'tJ -a: w 3: 0 a..

\

-40 I / 11 1/ 1/ ,

-50~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-15 -10 -5 o 5 10 15 e

Fig. 3-Curves showing the H-plane pattern at 6.6 GHz

Or-----------------------~--~----------·-------------,

-- SIMULATED

- - - MEASURED

-10

-2.0

-30

-40

_50~-~~--L-~-L--L-~-L--L-~-L--~~~--~~~·--~U-~

-10 -5 o

e Fig. 4--Curves showing the E-plane pattern at 10.65 GHz

1-:> 10

SHARMA & SRIHARSHA: SIMULATION OF RADIATION PATfERN OF MSMR ANTENNA 47

Or-----------------------~~----------------------~

-- SIMULATED

- - - MEASURED -10

-en -20 'tJ -a: w 3: 0 -30 Q.

\ (

-40 I

-50~~~--~~--~~~--~~--~~~--~~~~~~~~~~

-10 -5 o 5 10 e

Fig. 5-Curves showing the H-plane pattern at 10.65 GHz

O~------------------------~~--------------------------~

-- SIMULATED

- - - MEASURED

-10

-en -20 " -a: \ W 3: , 0 -30 Q.

(, /

-40

_50~~~~~~--~~--u-~~~~~--~~~L-~~~~~~~~

-10 -5 o 5 10

e Fig. 6-Curves showing the E-plane pattern at 18.0 GHz

48

-III ~ -c: W 3: 0 D..

INDIAN J RADIO & SPACE PHYS. FEBRUARY 1999

Or-----------------------~~----------------------~

-10

-III -20 ~ -c: w 3: o -30 D..

-40

-- 18HSIM.DAT

-- - - 18HH.DAT

I ... \ 1 \

1 \ 1 \ 1 \

_50~~~~~~~~~~~~--~~~-L~~~U~~~~~~

-10 -5 0 5 10

e Fig. 7-Curves showing the H-plane pattern at 18.0 GHz

1.0~--------------------------------------------------------·--~

-- SIMULATED

- - - MEASURED

-9.2

\

-19.4 I , r \

" \ \

-29.6 " \

-39.8

-50. 0 1I......L...L.IJ,..JI.LILI...L..L.~I..&./..I..I..L~~L.I___'_ __ L.._ ...... __'__u..........I..L.__'_U__Il.U..._L.l_A.u.uJ.LI.LI...I.L._..L_I -10 -5 o 5 10

e Fig. 8-Curves showing the E-plane pattern at 21 .0 GHz

SHARMA & SRI HARSHA: SIMULATION OF RADIATION PATrERN OF MSMR ANTENNA

Or-------------------------~~----------------------__,

-- SIMULATED

- - - MEASURED -10

-m -20 '0 -a: w ,.. 3:

, J \ 0 -30 I \ I a..

1\ I I \

\ I \/

-40

-50~~~--~~~~~~--~~--~~~~~~~~--~~~~~

-10 -5 o 5 10 e

Fig. 9-Curves showing the H-plane pattern at 21.0 GHz

50~-----------------------------------------------------,

40

30

20

m 10 '0 -a: w 3:

o

o -10 a.. -20

-30

-40

_50~wu~uu~~~~~~~uu~~wu~uu~~wu~~wu~uu~~

o 1 0 20 30 40 50 60 70 80 90 1 00 11 0 120 130 140 150 160 170 180

e Fig. 1 O-Curves showing the E-plane pattern at 18.0 GHz

49

50 INDIAN J RADIO & SPACE PHYS. FEBRUARY 1999

50

40

30

20

-CD 10 '0 -a: 0 w 3: 0 -10 a..

-20

-30

-40

_50~~~·U.~.u~~~~~~~~~~~~~~~~~~J~~u.~

o 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 e

Fig. II -Curves showing the H-plane pattern at 18.0 GHz

50

40

30

20

-CD 10 '0 --a: 0 w 3: 0 -10 a..

-20

-30

-40

_50~~~LUJ~~LUJ~~LUJ~~LUJ~~LUJ~~LUJ~~~~~LUJ~~

o 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180

e Fig. 12-E-plane pattern at 18 GHz without slope diffraction

SHAR ,1A & SRIHARSHA: SIMULATION OF RADIATION PATrERN OF MSMR ANTENNA 51

50

40

30

20

-m 10 '0 -a: 0 UJ == 0 -10 [l.

-20

-30

-40

-50U.~~~~~~~~~~~~~~~~~~~~~~~~~~~~

o 1 0 20 30 40 50 60 70 80 90 1 00 110 120 130 140 150 160 170 180

e Fig. 13-H-plane pattern at 18 GHz without slope diffraction

50

40

30

20

-m 10 '0 -a: 0 w == 0 -10 [l.

-20

-30

-40

_50~u.u.u.u.~u.u.u.u.u.u.u.~~~~~~~~~~~~~LULULU~

o 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180

e Fig. I4--E-plane pattern at 18 GHz without corner diffracti on

52 INDIAN 1 RADIO & SPACE PHYS, FEBRUARY 1999

50

40

30

20

-CD 10 "C -a:: 0 w 3: 0 -10 a..

-20 -

-30

-40 --50~~L~~~~~~~~~~~~~~~~~~~.~~~~~~~

o 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180

e Fig. 15-H-plane pattern at 18 GH z without corner d iffraction

Table I-Compari son of measured and simul ated - 10 dB beam effi ciency

Fn.:yuency Measured Simul ated G Hz effi ciency e ffi ciency

% % 6.6 87.35 87.17

10.65 90 .64 90.66

I X.O 90.43 90 .37

210 89. 11 88.63

Acknowledgements The authors wish to thank the Director, SAC,

Ahmedabad , fo r prov id ing them an opportunity to wo rk with Pro f. N Balakri shnan at lISe, Bangalore,

and for hi s encouragement and support. The authors al so thank Prof. N Balakri shnan, Chairman, SERC, lISe, Bangaiore, for agreeing to collaborate with SAC for development of software for reflector antenna in spite of hi s ' heavy schedule. The authors are also thankful to the ir colleagues in SAFG, for various help gi ven in carrying out thi s exerci se.

References I Collin R E, Antennas and rcrdiowave propagatioll (McGraw

Hill Book Co .. New York ), 1985, pp 244-253 . 2 Lo Y T & Lee S W, Antellna Handbook: Theory, Applicatioll

and Design (Van Nostrand Reinhold Co., New York), 1988. 3 Balan is C A, Advanced ellgineerillg electromaglletics (l ohn

Wiley and Sons . New Yo rk ), 1 989 , pp 743 ·850.