Effect of Fresnel zone clearance on propagation characteristics of microwave links in hilly terrains

(1)

b~I' ^)7,6

IndianJournalofRadio&SpacePhysics Vol.20,February1991,pp. 29-38

Effect of Fresnel zone clearance on propagation characteristics of microwave links in hilly terrains

C~~~:~Re~aVi,~&(M';~urthYDepartmentof Physics,SV University,Tirupati

'5'rr5~

.

'-

^...••.. ^-

~ Ci~l;;k;r: ~utta ~{:~eddy' '\

( RadioScienceDivision,NationalPhysicalLaboratory,NewDelhi110

oJV

/'-

The effect of Fresnel zone clearance on microwave\.

studied. The ray tracmg tediitique for radio rays t rou the troposphere has also been discussed.

The ray paths. were tra~~d for different ~levatio~ angles under varied..!!!.eteorologicalcondi~ojls. Un- der subrefractlon condItIons, there was msufficlent Fresnel zone clearance wtiicll leads to fade-outs of the received microwave signals. The antenna height at the transmitting end at Tiruttani was increased from 50 to 90 m to provide enough Fresnel zone clearance. A comparison of the performance of the microwave link before and after the modification of the antenna height has been presented. Propagation characteristics of Elagiri-Tirumala and Pallavaram-Tirumala microwave links situated over a hilly terrain are also presented. The field strength data have been analysed in terms of fade

rate, average fade depth and sCin~tio.n ;ndc><.The d;umal and ,earonal variation, of the above pammelers Me 01'0

d;"c"''';--'j ClJ rv.J-.

1

Introduction

\j

fraction varies with the geometry of the diffracting The terrain features of the geographical region object? Hall8 studied the diffraction losses for a are extremely important on two major counts path over which K, the effective earth's radius while locating a line-of-sight (LOS) microwave factor, is at a single point. Schiavone9 developed a link. First, the multipath fading is a dominant semi-empirical climatological model to predict the phenomenon that restricts reliability and data positive refractivity gradients in USA. Radio-path- transmission rates in high reliability links, espe- clearance measurement data in terms of Fresnel cially, where digital data transfer is involved^1-4. zone radius is given in CCIR Report](). In Austral- Secondly, any obstacle in the terrain, such as a ia, Harveyll developed a sub refractive fading hill, can obscure the signal path, if the clearance model using radiometeorological data. Experi- is short of the first Fresnel zone dimension. To ments12 on double knife-edge diffraction path add to this, if the region is located in warm, hu- showed that available theoretical/empirical meth- mid climate, the continuous and rapid change in ods are in good agreement with Deygoutl3 propa- the initial refractivity gradient in response to even gation model.

minor temperature variations will cause serious In general, whenever there is significant subre- fading and, in the extreme case, a total fade-out. fraction, the radio ray or the wavefront crosses While some of these effects can be alleviated by the obstacle at grazing incidence, even when the proper location of antennas, no mathematical for- optical LOS is clear. The field strength at the re- mulation can predict the design parameters and ceiving terminal is due to diffraction and the hence recourse has to be taken to empirical rela- transmission loss will be unacceptably large. The tionships developed from large samples of data extent of the clearance of obstacle relative to the collected over similar terrains. direct illS may be calculated in terms of "Fresnel Fading due to obstruction has been convenient- zone ellipsoids" drawn around the path terminals.

ly explained in terms of Fresnel zone radiis and This is the usual practice while planning the LOS first Fresnel zone6• It is found that loss due to dif- links over hilly terrain. However, this is not possi- 29

(2)

INDIAN J RADIO & SPACE PHYS,FEBRUARY· 1991

ble partly because of lack of such statistics and partly because the antenna heights become ex- cessively large. A general guideline for wet climates is normally given by considering K=1,and the value of K is kept as low as 0.5 for desert areas. One of the serious problems with high humidity climates is that even small variations in temperature affect the refractivity gradient in a major way. Subrefraction can occur for several hours when the humidity gradient totally vanishes, even if the humidity itself is high. This could be further accentuated when a cold wind mass is car- ried over the hill,and thus increasing the temperature lapse rate. To summarize, an illS link over a hilly terrain should be so designed that first Fres- nel ellipsoid, which essentially gives rise to coher- ent phase contribution at the receiver, should be above the obstacle for the percentage of time dic- tated by the planned reliability of the link. In order to investigate if the observed fade-outs for Ti- ruttani-Tirupati microwave link could be due to insufficient obstacle clearance, an attempt has been made to find out whether first Fresnel zone clearance is secured along the entire propagation path under varied meteorological conditions in the month of March during the early hours of the day when, generally, the fade-outs are observed in the said microwave link.

2 Terrain characteristics

The LOS microwave links between Tirupati-Ti- ruttani, Elagiri-Tirumala and Pallavaram-Tirumala are situated in hilly region, the terrain characteristics of which are given in Table 1. The system characteristics of microwave links are given in Table 2. Fig. 1 shows the path profile for the three microwave links. While drawing the path profiles the value of K is taken as 1.33 which ref- ers to normal refraction. From the terrain profiles it is evident that for Elagiri-Tirumala and Pallava- ram-Tirumala microwave links the rays reaching

Table 1 - Terrain characteristics of different microwave links Microwave

link

Path length

krn

Transmitting antenna

height m

Receiving Recording

antenna end

height m

Tiruttani- 60 90 950 Tirupati

Tirupati

Elagiri- 142 llOO 1020 Tirumala Tirumala

Pallavaram- 117.5 160 1100 Tirumala

Tirumala

30

the antenna have sufficient Fresnel zone clearance, as there are no obstacles on the ray path.

However, for Tiruttani-Tirupati microwave link, some of the major hillsare at a distance of 20, 35 and 42 km from the transmitting end [Fig. l(a)].

The noticeable feature is a hillof 400 m height at a distance of 20 km from the transmitter and 40 km from the receiver. In the month of March during the early hours of the day, severe fadings of the order of 30-40 dB were observed.

3 Fresnel zone andits clearance

The propagation region bounded by the first Fresnel zone for Tiruttani-Tirupati LOS microwave link is shown in the Fig. l(a). The critical point along themicrowave linkis hill top 1.

The radius of the first Fresnel zone at the hill top 1 is given as"

... (1)

where II is the distance between the transmitter and hill 1, '-2 the distance between the receiver and hill 1, A the wave length and hi the radius of the firstFresnel zone.

From the situation shown in the Fig. l(a), we have, II = 20 km, 12= 40 km, A= 4 em, and hence hi =23 m.

When a radio ray passes through troposphere it encounters the refractivity variations in the troposphere which, in turn, affect the ray due to various propagation mechanisms, viz. atmospheric refraction, reflection from and trapping in layers, turbulent scattering, diffraction by smooth sur- faces etc. There are various techniques to evalu- atethe bending of the ray.

Assuming a model of monotonic decrease of refractivity for the troposphere, an empirical relation is given to evaluate the refraction errors and bending. However, the degree of approximation is high in this empirical relation.

4 Ray tracing technique

The basic numerical ray tracing technique ^14-16 combined with complex propagation loss comput- ations by numerical means is the most powerful tool and has almost replaced the normal mode.

Troposphere is considered to be stratified into spherical layers of thickness ~ and constant refractive index ^fly"^, so that the vertical gradient of

fly" isassumed to be constant withineach slice.

It might appear that radio waves are propagat- ed in the same way as they do over a smooth surface,and instead of a single reflected ray which is formed over a smooth surface there might be

(3)

Table 2 - System characteristics of different microwave links

System characteristics Link

Tiruttani-Tirupati Elagiri-Tirurnala Pallavaram-Tirumala

7.659 7..394 7.394

Parabolic dish Parabolic dish Parabolic dish with horn feed with horn feed with horn feed

40 45 46

Parabolic dish Parabolic dish Parabolic dish with horn feed with horn feed with horn feed

40 45 46

Frequency (GHz) Transmitting antenna

Gain oftransmitting antenna (dB) Receiving antenna

Gain of receiving antenna (dB) Transmitted power (Watt)

Recorder time constant (msec) 300 300 300

b

••-TRANSMIlTING ANTENNA, TlRUTTANI b- RECEIVINGANTENN••••

TlRUPATI 1 -ANNALAYA GUTTA 11·NARIGAOIKONOA

tu-SATARAVADIKONOA

0~~---~~~----'4nO---VJO'---~2OnO----~'---~

DISTANCE, Km

16 E

20 40 60 80 100 120 140 DISTANCE Jkm

(142 km) TlRUMALA

ELEVATlON:I100m ANTENNAHEIGHT·15m

ELAGIRI ELEVATION.1020m ANTENNAHEIGHT.80 m

---

-r--r---

K - r--r--r----

I-- ~

----

^I---

---

^_______

rz

"--.."---

---..

',,"--- - r---____..---

-f- <,

- ---r--I---

I--f-- r-, ^----I--- ^'---..

'--" ^<, ~

---

-

¹ ^~r--- -->---...,

^r---

~ ^t- (\ ..it

~::S:---

f-- '-" \. ../ \

L--- <,

-~'~~

0- 20 40 so ₈₀

'-~21

¹⁰⁰ ^-

1200

1000

800 E Z 600 Q

>-

..

>

~ 1.00 w

OISTAN ca,km

Fig. 1 - Terrain profile of the microwave link between (a)Ti- rupati and Tiruttani, (b)Elagiri and Tirumala, and (c)Pallava-

ram and Tirumala

several reflected rays coming from points where the angle of incidence is equal to the angle of reflection. But this is not so, because the reflected ray is formed within an area bounded by the first Fresnel zone and not at a geometrical point, and in most cases the tops of the hills are much small- er in size than the first Fresnel zone". Thus a smooth convex earth is assumed for ray tracing of radio waves in a hilly terrain.

To trace a radio ray from the point 1 on the earth's surface to the point 2 at a height above the surface, the geometry'? isas shown in Fig. 2.

Assuming earth's radius to be 6370 km, the sea level arc distance is found to be (6370lt)d'2' where d12 is the distance between points 1 and 2, and subtends an angle <l»at the earth's centre. Let the ray make an angle

a,

with the horizontal at one end, O²at the other end and undergo bending through an angle ^'1:'2' Further, let n, and nz denote the values of refractive index corresponding to r and '2'respectively. Then, we have

... (2)

and

r:

⁼ n,'1 cosB,

f

^dr/r{n^2/ ^- ni,icos²

0r

¹¹² ^... ⁽³⁾

... (4)

120

Eqs (3) and (4) give the fundamental relation-

ships needed for tracing the radio rays.

Thus for very low elevation angles (

<:

10°), ray bending maybe obtained from Eq. (2)as

(5)

(4)

INDIAN J RADIO &SPACE PHYS, FEBRUARY 1991

"2

Radio fay

Fig. 2 - Geometry of radio ray refraction

spherical coordinates r, ,y, with an undefined independent variable, ^I,are as follows.

~(j!) _

^dl ^ds dn" ds _ [nr!^{dr dl} d<l>2_ Sin2<1>(dy)2]jds^dl ^dl ^dl

=

0 .. , (8)

d ( 2d<l» dn dS[ 2. (dy)2] 'Ids dl nr ~ - d<l>"dl nr smcosdl Idl=O

... (9) ... (10)

The independent variable, I, can be chosen from a number of workable variables to give different forms for computation. Here it is identified with the distance along the ray. The refractive index is now restricted to be a function of r and

only, and the axes are chosen to be geocentric with the transmitter at <1>

=

O.Then Eq. (10) states that the rays stay in longitudinal planes. The following variables are introduced for computational converuence.

where ds is an elemental distance along the ray dDlds= re( V7nr) and denotes the variation of the integral. The re-

fractive index, n, is real and is obtained from the dUlds=( VZlnr')+ dn/dh temperature, pressure and water vapour pressure,

and is given by the following equation. dhlds= U/n where lltl, /(J, 80 are the coordinates (refractive in-

dex, radial distance and elevation angle, respectively) of the initial point on ray path and n], r1, 8]

are the coordinates of the subsequent points.

For low elevatiQn angles ^« 10°), i.e. for near horizon propagation, ray description of propagation is not sufficient and recourse has to be made to the wave propagation.

Fermat's principle of stationary phase-transit time can be written as

... (11) '" (12) ... (13) ... (14)

... (15) ... (16) ... (17) '" (18) v= (nr Ire)X(dDI ds)

U= ndhlds

where D is the range, h the height coordinate and ^rethe radius of the earth.

Eqs. (8) and (9) then become .,. (6)

a

f n.ds=U

path

(n - 1)X 106

=

77.6(PIT)

+

3.73 X 105(effl) ... (7) where,

P= Atmospheric pressure in inbar T= Temperature in K

e=Water vapour pressure in mbar

The Euler equations derived from Eq. (6) m

If a ray starts at the surface of the earth at an angle

a

to the horizontal, the initial conditions then become

h=O,D=O,dDlds=cosa,dhlds=sina ... (19) Ray path may be obtained by numerically inte- grating the differential Eqs. (15 )-(18).

\'"-

32

(5)

\,1

5 Ray.paths using radio refractivity index gradients

Radiosonde observations taken at Madras by the India Meteorological Department were used to obtain the refractivity profiles. Madras is at a distance of 50 km from the transmitting end, Ti- ruttani. This is the only radiosonde station near Tiruttani-Tirupati microwave link. Generally the climate at Tiruttani and Tirupati is the same as that of Madras and hence the radiosonde data of Madras were used to build the refractivity profiles. Some typical refractivity profiles observed during March are presented in Fig. 3.

A computer programme has been developed using the ray tracing technique!7. The path of the radio ray propagating at 7.6 GHz was traced under different meteorological situations using the ray tracing technique. The exercise was done by changing the elevation angle ftom 0.1. to 1.2° in steps of 0.1°. It was found that the rays reach the receiving end when the angle of elevation is just around 0.8°. In other cases, the radio rays did not reach the receiving end. The rays were found to propagate over the receiver when the angle of elevation was more than

O.r,

while for elevation angles less than 0.8°, the rays were found to pass below the receiving antenna. The ray paths traced for elevation angles of 0.1° and 1.0° are presented in Fig. 4. Fig. 5 shows the ray paths traced under different atmo~pheric conditions when the refrac-

tjvity gradi@nt is of higher order. Fig. 5(a) illus- trates the ray path when the initial refractivity gradient is - 20 N-uriitlkm. It depicts that the rays having an elevation angle of 0.8° get obstructed by the hill under this atmospheric situation. Fig. 5(b) presents the ray path for an atmospheric situation when the refractivity gradient is of

- 40 N-unitlkm. During this atmospheric condition the rays are found just to clear the hill top.

Figs 5(c) and 5(d) present the paths of the rays when the atmosphric condition is super-refractive having initial "refractivity gradients of - 75 and

-130 N-unit/km, respectively.

It can be observed from Fig. 5(a) that the rays are obstructed by the hill top 1 under sub-refractive conditions when the refractivity gradient is less than - 40 N-unitlkm. It cah also be observed from the typical refractivity profiles (Fig. 3), that sub-refractive' conditions prevail in the month of March when, generally, severe fade-outs are observed in the early hours of the day. Fig. 5(a) also shows that there is no Fresnel zone clearance at hill top 1 under sub-refractive conditions ^11. It may, therefore, be inferred that the observed fade-outs are due to insufficient Fresnel zone clearance at the hill top, particularly, during the sub-refraction conditions. Thus, to provide sufficient Fresnel zone clearance even under sub-refractive conditions, the antenna height at the transmitting end, i.e. Tiruttani, is increased from 50to 90m.

OOOOhrs GMT

26-3-'987

250300 350

RE F RACTIVI1Y, N-units 200

OOOOhrs GMT

3-3-1987

REFRACTIVITY, N-units"

MADRAS

o

200 2·5,

2·0

~ ,·5

.

UJ ^C

UJ

;:)

C

•...

;:)

•... •...;;. ',0

;:: ,·0 -oJ<C

O.J ^0·5

(

Fig. 3 -Typical R.RI profiles observed over Madras on 3-3-87 and 26-3-87 (0000 hrs GMT)

33

(6)

INDIAN J RADIO &SPACE PHYS, FEBRUARY 1991

RAN G E,km

Fig. 4 - Ray paths traced for elevation angles of 0.1 and 10

100Qi

50

TIRUTTANI-T1RUPATI LOS 7"6 GHz

height was increased. Table 3 shows a comparison of amplitude variation of signal level at different probability levels.

6.2 Fade rate

Fade rate is defined as the number of fades in a particular time interval and usually determined by counting the number of intt:;rsections of the median signallevel·in a specified time interval.

From the cumulative distribution of fade rates, it was seen that at 20% probability level the fade

Table 3 - Amplitude variation of signal levels at different levels of probability

TIRUTTAN)-TIRUPATI ,LOS7-6GHz

Probability level

%

Amplitude before Fresnel zone

clearance

Amplitude after Fresnel zone

clearance

AN:.-20

2'Ju

D_o '0

20 10 5

Median signal level (dBm) exceeded

-24 -40

-48

-16 -19 -28

wo

"

~^,

2')0

40

AN:. ~IJO

20 10 5

Fade rate exceeded 5 8 15

2 6 12

RAN G E • km

Fig. 5 - Ray paths traced under an atmospheric situation of:

(a) N= -20 N-units/km, (b) -40 N-units/km, (c) -75

N-units/km and (d) - 130 N-units/km 20

10 5

Fade depth (dB) exceeded 3

5 10

2 3 5

Scintillation index (%)exceeded ••

20

6 3 10

176 5

5231

Fade-out time (min.) exceeded 20

28 23 10

47 35 5

5345

6.1 Median signal

The signal level which exceeded for 50% of the total valid recording time is known as the median signal. From cumulative distribution of the hourly median signal level for Tiruttani-Tirupati micro-

wave link, it was found that at 20% of time the Note: Total time of fade-out observed before and after the

variation in the signal level was above - 24 dB, Fresnel zone clearance was 44 and 29 hr, respectively.

whereas it was above

-16

dB when the antenna 6 Improvement of performance of microwave

link with adequate Fresnel zone clearance The microwave field strength has been re- corded at Tirupati in March 1987 after the antenna height at the transmitting end is increased. A comparison of the performance of the microwave link is made before and after the antenna height is changed.

34

(7)

rate was more than 5 per hour, whereas it was 2 fades/hr when the Fresnel zone clearance was provided by increasing the antenna height at the transmitting end. At different probability levels, the comparison of fade rates is shown in Table 3.

6.3 Fade depth

The irregularities in the atmosphere along the path of the radio rays affect the velocity of propagation and causes difference in phase which, in turn, produces fading. Usually fading is expressed in terms of fade depth. Fade depth is defined as the difference between maximum and minimum signal variation over a small interval of time.

From cumulative distribution of the average fade depth for Tiruttani-Tirupati microwave link, it was found that at 20% of time, the average fade depth is more than 3 dB and after the antenna height is increased, it was only 2 dB. At 10 % of time, the average fade depth was 5 dB and after the modification, it was only 3 dB. Similarly at 5% of time the average fade depth was 10 dB and after the antenna height is increased, it was 5 dB. Table 3 depicts variation of average fade depth at different probability levels, before and after the antenna height is increased.

6.4 Scintillation index

The scintillation index (SI) is defined ^aslX

S.l. (%)

=

Pm"" - PminX 100

Pmax

+

^Pmin

where, ^l~l1"x is the power of the third peak down frOm the maximum excursion of the signal

strength and ^Pmin is the power of the third minimum upwards from the minimum signal strength.

From the cumulative distribution of SI for Ti- ruttani-Tirupati microwave link, it is observed that at 20% probability level, the SI is above 6%, whereas it is 3% after the modification (Table 3).

At 10% probability level, the scintillation index is above 17%, whereas it is 6% only after modification. Similarly at 5% probability level, the SI is above 52%, whereas it is 31% after the antenna height is increased.

From the cumulative distribution of total fade- out time (in min), it is seen for Tiruttani-Tirupati microwave link that at 20% of time (probability level) the fade-out is for 28 min and after the antenna height modification it is for 24 min only (Table 3). At 10% probability level, total fade-out time is for 47 min and after the antenna height is increased it is for 35 min only. Similarly at 5% of time, total fade-out is for 53 min, whereas it is 45 min after the modification.

The total fade-out time of the signal has been computed for Tiruttani-Tirupati link in March 1987 after the antenna height is increased. It was observed that for 29 hr the signal was completely faded out in March 1987, whereas the total fade- out time in March 1983 (before the modification) was 44 hr as shown in Table 3.

7 Propagation characteristics of Elagiri-Tirumala and Pallavaram-Tirumala microwave links Cumulative distributions of hourly median basic transmission loss for Elagiri-Tirumala and for Pal- lavaram-Tirumala microwave links are shown in Figs 6 and 7, respectively.

17

)·0 J~O

PERCENTAGE OF TIME ORDINATE EXCEEDED 220

CD

-a 210

</l

o

...J

Zo

gj~

</l

Z

<l:

a:>-

"" 180

</l

<l:

CD

161

01

"".

~-~

LOS ·7·]

p.AAE~""ONSOON W: WINTER

M:MONSOON

0'POST-f040NSOON p

VI

Fig. 6 - Cumulative distribution of hourly median basic transmission loss for Elagiri-Tirumala microwave link

35

(8)

INDIAN JRADIO &SPACE PHYS, FEBRUARY 1991

22

PAllAVARAW _ T1RUWAl A

~11.~

P'PIlE-WOHSOQN

•• :WOHSOON

<D

" 210

P

170 In(/l

o...J

Z 200

o

(/l(/l

~1z

~I-

u 180

Vi

<{

<D

1~.i- .--- ---···---Hl-- --

^100{) ^100-0

PERCENTAGE OF TIME ORDINATE EXCEEDED

Fig. 7 - Cumulative distribution of hourly median basic transmission loss for Pallavaram-Tirumala microwave link

Fig. X- Diurnal and seasonal variations of fade rate observed for Elagiri-Tirumala microwave link for different seasons

pheric conditions that prevail in these hours. The nighttime high fade rate is due to the multi path fading occurring as a result of layered structure of the atmosphere. The low fade rate observed during the daytime is due to the well mixing of the atmosphere.

The diurnal variation of average fade depth observed during all seasons for Elagiri-Tirumala microwave link is shown in Fig. 10. The average fade depth is maximum in pre-monsoon followed by winter, post-monsoon and monsoon season.

The maximum fade depth is observed in the transition hours (0700-1000 hrs) of pre-monsoon season. Fig. II shows the average fade depth for pre-monsoon and monsoon seasons for Pallava- ram-Tirumala microwave link. The fade depth is

"

1 1M E,hrS IST

j fII

11- "

1 ~ 1IIIiII lXJU(j PO",!-f.o()NS0')'<

;JeS-e'

JJ-Ttf JPl

~ ' I r11

, ,

,:~~,Cl.illp

J, 1200

::.12

From Fig. 6 it can be seen that at 70% probability level, for pre-monsoon season, the transmission loss exceeds 167 dB, for winter it exceeds 168 dB, for monsoon it exceeds 166 dB and for post-monsoon season it exceeds 165 dB. Similarly at 90% probability level, transmission loss for pre-monsoon, monsoon, post-monsoon and winter seasons exceeds 177, 169, 168 and 172 dB, respectively.

Fig. 7 shows that in pre-monsoon season, at 70 and 90% probability levels, the transmission loss exceeds 168 and 187 dB, respectively. Fig. 7 also shows 'that for monsoon season the transmission loss exceeds 165 and 167 dB, respectively at 70 and 90% probability levels.

Fig. 8 shows diurnal variation of fade rate observed during all seasons for Elagiri-Tirumala microwave link. From Fig. 8 it can be seen that for pre-monsoon, the observed fade rate is high as compared to that for winter, post-monsoon and monsoon seasons; and also in the transition hours (0700- ¹⁰⁰⁰ hrs), the fade rate is maximum for all seasons. For all the seasons fade rate is minimum from 1000 to 1700 hrs. The fade rate observed over this microwave link varies from 0 to 500 fades per hour.

Fig. 9 -shows diurnal variation of fade rate observed during pre-monsoon and monsoon seasons for Pallavaram-Tirumala microwave link. The observed fade rate varies from 0 to 74 fades per hour. The maximum fade rate is observed during pre-monsoon season. The high fade rate observed for this microwave links in the morning (i.e., in the transition) hours is due to the unstable atmos-

(9)

NARAYANA RAO ^{et aL:}FRESNEL ZONE CLEARANCE EFFECT ON MICROWAVE PROPAGATION

1

ⁱ^I

Fig. 13 shows the diurnal variation of the scintillation index for Pallavaram-Tirumala link for pre-monsoon and monsoon months. In the pre- monsoon season the maximum scintillation index is 15%, whereas in the monsoon months the max- imumis 13%.

For Pallavaram-Tirumala microwave link, the post-sunrise and post-sunset periods are characterized by high values of scintillation index. Sun- rise and sunset periods are characterized by the maximum in pre-monsoon as compared to that in

monsoon season.

The atmospheric variability during pre-monsoon and winter seasons is higher than that in the monsoon and post-monsoon seasons. Fig. 12 shows monthly variation of refractivity gradient

1i.N for 0000 and 1200 hrs GMT for the year 1987. The variation of average refractivity gradient in pre-monsoon and winter months is - 30 to - 85 N-units/km, while in monsoon and post- monsoon seasons the gradient values vary between - 50 and - 80 N-units/km. The results in- dicate that the atmosphere is well mixed in monsoon and post-monsoon seasons.

PALLAVARAM -TIRUMALA

51- Las. 7GHz

PRE -MONSOON

PALLAVARAM -TIRUMALA

BOr LOS 7 GHz PRE-MONSOON co

-0

0600 ^??oo

60

40

~

.c:

---

<J)

w

o

.•• 30 lL

0600

MON SOON 000(1

I

>----

"-

wo

...

lL

1200 TIME hrs,lST

MONS00N

20

10

Fig. II - Diurnal and seasonal vanatlon of fade depth observed for Pallavaram-Tirumala microwave link for pre-mon-

soon and monsoon seasons

100

PRE·MONSOUN

1911~-lll

f,If' J J A SON 0

MON TH S

Fig. 12 - Monthly variation of refractivity gradients observed during the year 1987

90;

E

---... 80'

.~ ^Z^::l^..:._~_z>----" ⁷⁰

z ^60'

UJ ^II:^Clc^c( 50'

>- t:^>----~u.. ⁴

II: lLUJII:

3D

WINTER 198"HI'

"

Pl)";l_MUNSOON '\111'111'

~ "r---

" '

<

1 rr\h Ill] lnp

"

,

"'

Fig. 10 - Diurnal and seasonal vanatlons of fade depth observed for Elagiri-Tirumala microwave link for different sea-

sons

TIME,hrs 1ST

Fig. 9 - Diurnal and seasonal variations of fade rate observed for Pallavaram-Tirumala microwave link for pre-monsoon and

monsoon seasons

37

(10)

I

INDIAN JRADIO &SPACE PHYS, FEBRUARY 1991

Acknowledgement

The authors gratefully acknowledge the Nation-

al Radar Council, Department of Electronics (DOE), Government of India, for sponsoring the research project "Studies on Anomalous Radar and Microwave Propagation in Hilly Terrains".

The present work forms a part of the project. The authors (KKR, KSR and SVBR) are also thankful to DOE, for providing them research fellowships.

Thanks are also due to the General Manager and the Director of Southern Telecommunications Re- gitln, Madras, for giving permission and providing necessary facilities to carry out the observations at their microwave station at Tirupati. The authors wish to express their sincere thanks to the Director, India Meteorological Department, Mad- ras, for providing radiosonde data. They are also thankful to the authorities of Sri Venkateswara University, Tirupati, for providing necessary facilities to carry out the research work.

References

1 Bullington K, Bell Syst TechJ( USA), 50 (1971) 2039.

2 Barnett W T, IEEE Trans Commun (USA), 27 (1979) 1842.

3 Greenstein L J, IEEE Trans Commun (USA), 26 (1978) 1247.

4 Stephansen E T, IEEE Trans Commun (USA), 27 (1979) 643.

5 Bullington K, Bell Syst Techi( USA), 36 (1957) 593.

6 DoJukhanov M, Propagation of Rudiowaves (Mir Publish- ers, Moscow), 1971. 122.

7 White R F, Engineering considerations for microwave communication systems (Lenkust Electric Co., San Carlos, California), 1970.

8 Hall M P M, Effects of the troposphere on radio commu- nication (Peter peregrinus Ltd., ITE electromagnetic waves series H,London), 1979.

9 Schiavone J A, Bell Syst Techi( USA), 60 (1957) 803.

10 CClR Repon, 339-4,1982.

II Harvey R A, ILl:'/:' hans Amennas &Propag (USA), 35 (19H7),H32.

.12 Tewari R K, Jassal B S & Roy M N, Indian

i

^Radio ^&

Space Phys, IH (1989) 14.

13 Deygout J, ILI:"I: Trans Antennas & Propag (USA), ¹⁴ (1966)4HO.

14 Inston H H&Curtis A R, Radio Sci( USA), 3 (1968) 27.

1'1 Wehsll'r i\ R, IF/:F hallS AllIell/wl' & I'ropag (U.~A), 30(1982)796.

16 Bean B R & Dutton E J, Radio meteorology (Dover pu- blication Inc., New York), 1966, 165.

17 Rao D N, Kesava Murthy M J, Sarkar S K, Dutta H N &

Rl'ddy B M, II:!:F hum ,11/1l'IIl/til &I'wpul!, (U.~A): 35 (19H7) 1330.

)H Whitney H E, Malik C & Aarons J, Planet &Space Sci (GB), 17(19691 106<).

0000

0000 MONSOON

1800 PRE - MONSOON

1200 0600

PA LLA VARAM- TIRUMALA

LOS 7 GHz

10

.•• 5

15

><

UJa z z

~ 1~

>--

<-'

:: 10

>--

z

~ 6

TIME,hrs 1ST

Fig. 13 - Diurnal and seasonal variation of scintillation index observed for Pallavaram-Tirumala microwave link for pre-

monsoon and monsoon seasons

dissolution and formation of nighttime temperature inversions, a common feature in the diurnal cycle of the boundary layer. Also, in the nighttime, high values of scintillation index are observed due to the temperature inversions in the atmosphere. Scintillation index is low and steady during the daytime when the atmosphere is con- vective.

8 Conclusions

The comparison shows that there is an improvement in the median signal level, fade rate, average fade depth, scintillation index (per cent) and signal fade-out (of time) for Tiruttani-Tirupati microwave link after the height of the transmitting antenna is increased by 40 m. Particularly at 5%

probability level the improvement is significant. A significant decrease in the total fade-out time is also seen after the modification. From the above analysis it is concluded that the improvement is due to the first Fresnel zone clearance along the entire propagation path which is secured hy increasing the antenna height at the transmitting end.

38

Effect of Fresnel zone clearance on propagation characteristics of microwave links in hilly terrains

b~I' )7,6