NTIA Report 90-256

Design and Performance of a Long,

Over-Water Microwave Radio Link

J. E. Farrow

U.S. DEPARTMENT OF COMMERCE

Robert A. Mosbacher, Secretary
Janice Obuchowski, Assistant Secretary
for Communications and Information

February 1990
 CONTENTS

Page

LIST OF FIGURES iv

LIST OF TABLES v

ABSTRACT 1

1. INTRODUCTION 1

2. DESCRIPTION OF THE LINK .' 1

2.1 The Area . . . . 2
2.2 The Profile and Link Antenna Layout 2
2.3 Radio Equip~ent Used 2

3, DATA GATHERED , . . . 3
3.1 Recording Setup , . 3
3.2 Strip Charts 3

4. DATA ANALYSIS . . . 3
4.1 Signal Level Calibrations 3
4.2 Propagation Regimes 6
4.3 Analyzing the Strip Charts 8

5. PRESENTATION . . . . 8
5.1 Comparison of Predicted and Measured Signal Level Data 8
5.2 Long Term Signal Level Effects 9
5.3 Anecdotal Analysis 10

6. CONCLUSlONS AND RECOMMENDATIONS 18

7. ACKNOWLEDGMENTS 18

8. BIBLIOGRAPHY 21

APPENDIX: LONG-TERM LINK RELIABILITY PREDICTIONS 23

A.1 The Rain Attenuation and Multipath Analysis 23
A.2 The Obstruction Fading Analysis. .. 23
A.3 Discussion 24
A.4 References 25

iii
 LIST OF FIGURES

Page

Figure 1. Plan view of the She11man-B1uff-to-Ocean-Tower radio

link. 4
Figure 2. Link diversity configuration. 5
Figure 3. Link received signal level recording setup. 7
Figure 4. Propagation regimes by time, March 1989. 11
Figure 5. Propagation regimes by time, April 1989. 12
Figure 6. Propagation regimes by time, May 1989. 13
Figure 7. Propagation regimes by time, July 1989 14
Figure 8. An example of quiescent propagation conditions, regime
1. 15
Figure 9. An example of the transition from propagation regime
1 to regime 2. 16
Figure 10. Regime 3, bottom antenna showing the highest
signal. 17
Figure 11. Regime 4, middle antenna showing the highest
signal. 19
Figure 12. Regime 5, all antennas. showing severely depressed
levels. 20
Figure A-l. Path profile for the mu1tipath attenuation
distribution calculation. 28
Figure A-2. Multipath attenuation distribution. 29
Figure A-3. Link pre-detection carrier-to-noise ratio
distribution. 32
Figure A-4. Measured refractivity gradient distribution. 33
Figure A-5. Calculated bi-moda1. refractivity gradient
distribution. 35
Figure A-6. Ray path for probability of 0.5 corresponding to
k-1.472. 36
Figure A-7. Ray path for probability of 0.1 corresponding to
k - 1.071. 37
Figure A-8. Ray path for probability of 0.01 corresponding to
k - 0.709. 38
Figure A-9. Ray path for probability of O. 001 corresponding to
k - 0.544. 39
Figure A-10. Ray path for probability of 0.0001 corresponding to
k - 0.454. . 40
Figure A-11. Ray path for probability of 0.00001 corresponding to
k - 0.370. • 41
Figure A-12. Mixed mode combined long-term path loss
distribution. 43

iv
 LIST OF TABLES

Page

Table 1. Summary of Predicted Link Performance . . . . . . 9

Table 2. Signal Level Regimes by Fraction of Time . . . . . 9
Table A-1. Path Geometry Input and Calculation Output Table. 26
Table A-2. Meteorological Parameter and Basic Transmission Loss
OUtput Table. . 27
Table A-3. Carrier to Noise Probability Distribution
Parameters.. . 30
Table A-4. Digital Link Performance Parameters. . . 31
Table A-5. Bi-ModalRefractivity Gradient Distribution
Parameters. . . . . . . . . . . ... 34
Table A-6. Dominant Mode Determination and Transmission Loss
Dis tribution. . . . . . . . . . 42

v
 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
.I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
 DESIGN AND PERFORMANCE OF A LONG, OVER-WATER MICROWAVE RADIO LINK

J .. E. Farrow*

This paper discusses the design and the measured performance

of a 76.8-kIn (47.7-mi) over-water 2.3 GHz microwave radio link.
Since one of the terminals was located on an ocean tower, the
maximum antenna height at that location was restricted to 49 m
(160 ft). A 367-m (1200-ft) tower was used at the shore end to
provide ray path clearance for normal and extreme refractivity
gradients. The tower at the shore end supported antennas at 361 m
(1180 ft), 345 m (1128 ft), 229 m (750 ft), and 49 m (160 ft) to
provide quadruple diversity protection. Received signal level
measurements were made over the link for several months including
some of the summer period. Propagation performance of the link
during this period was satisfactory.

Key words: 2 GHz propagation; long; over-water path; quadruple diversity

performance analysis; refractive gradient analysis

1. INTRODUCTION

This long, over-water link supports the operation of the U. S. Navy'·s

Charleston Tactical Air Crew Training System (TACTS) located off the coast of
Georgia, south of Savannah. It connects a group of off-shore communication
platforms with the land communication links at Shellman Bluff, GA, which carry
the information signal to the Marine Corps Air Station at Beaufort, SC. The
Naval Air Test Center at Patuxent River, MD, requested the aid of the Institute
for Telecommunication Sciences (ITS) in designing a link-diversity structure
that would meet the Defense Communications Agency reliability requirements for
digital traffic to be carried over the link [Kirk and Osterholz, 1976]. The
considerations submitted to the Navy are outlined in the Appendix. The
Institute was then asked to assist the Navy in validating a contractor's
suggested implementation, and then to consult with the Navy in evaluating the
actual link performance.

2. DESCRIPTION OF THE LINK

This microwave radio link at the Charleston TACTS range provides a 6 Mb/s
connection between an ocean tower and the shore facilities. It operates in the
1.7-to-2.4 GHz Government band allocated to fixed point-to-point service. This
frequency band was chosen since the rain attenuation at these frequencies will
not be a limiting factor in link reliability [Crane, 1980]. Although the link
is only 76.8 kIn (47.7 mi) long, the height restriction on the ocean tower
antenna required a 367·m (1200-ft) tower at Shellman Bluff to provide adequate
path clearance. Since the link is over water and since such an aSYmmetrical

* The author is with the Institute for Telecommunication Sciences, National

Telecommunications and Information Administration, U. S. Department of Commerce,
Boulder, Colorado 80303.
antenna arrangement is necessary, special attention was paid to the design of
the diversity scheme to assure propagation reliability.

2.1 The Area

Figure 1 is a map showing the general area of the link and its
relationship to the coastline. Because the area of coastal Georgia has been
used extensively as an experimental area to establish and validate propagation
models, the area was known to provide a challenging environment for such a link.
[Vigants, 1975; Barnett, 1979]. Other options for the link, such as underseas
fiber optic cable, were considered, but the microwave radio option was chosen
because it promised to provide nearly the required level of service
(see Table A-4) at the most attractive cost. The lack of an intermediate relay
site required that the Navy accept the 76.S-km link and devise an equipment
configuration that would provide the propagation reliability desired.

2.2 The Profile and Link Antenna Layout

The link profile was easy to construct since the shore station has an
altitude of 3 m (10 ft) and is within a few kilometers of the shoreline and the
ocean tower is located in about 20 m (66 ft) of water. The profile diagram
shown in Figure 2 includes the towers and the antenna locations. Note that for
normal refractive conditions (k - 1.42) the bottom antenna is well below the
radio horizon. The placement of the antennas reflects the same design
philosophy used in the design of a Defense Communication System microwave link
across' the English Channel [Zebrovitz, 1975; Wortendyke et a1., 1979]. The
purpose of the physical arrangement in each case was to design around the known
propagation features of the sea-land interface and over-water links to provide
transmission continuity to the maximum extent possible.
. Basically, the highest antenna was located to provide clearance over the
curve of the Earth for as small an effective Earth's radius factor as possible.
The second antenna was located 16 m (53 ft) below the top antenna to provide
good space-diversity protection.against correlated phase interference fading
[Vigants, 1975]. The mid-level antenna was located to provide 1ine-of-sight
reception for normal conditions for widely spaced diversity reception, and the
bottom-level antenna was located to take advantage of the signals propagated
through surface atmospheric ducts [Dougherty, 1969].

2.3 Radio Equipment Used

The radio equipment selected for this link was commercial quality digital
microwave transmitters and receivers with the exception of the receiver
diversity combiners. To take full advantage of the elaborate diversity scheme,
four-fold (quad) diversity combiners developed for use on tropospheric scatter
systems were procured and installed. This combiner is of the type that
selectively amalgamates the four IF signals (a quad diversity, equal gain, pre-
detection combiner) in proportion to their individual signal-to-noise ratios to
produce a single signal to deliver to the detector/demodulator. Such a device
will produce very nearly the optimum signal-to-noise ratio from the combination
of the four available signals [Brennan, 1959].
A 6.2-Hb/s traffic baseband and a 192-kb/s orderwire and control channel
bit stream are combined in the radio, and this composite stream imposes

2
four-phase modulation on the carrier. This modulation scheme at this bit rate
is expected to be relatively immune to the effects of amplitude nonlinearities
that are caused by phase-interference fading [Hause, 1981]. Table 1 shows the
contractor's calculations for link equipment requirements and the resulting
propagation performance. The 45-dB dispersive fade margin shown in Table 1
supports this contention since a greater vulnerability to nonlinear fading would
be reflected in a somewhat lower (30 to 40 dB) dispersive fade margin.

3. DATA GATHERED

The purposes for recording the received signal levels (RSL's) for this
link were to validate the design of the diversity system and also to test the
installation for acceptance by the Government. Although the diversity design
and the guarantee of link performance were largely the responsibility of the
Government, the contractor was responsible for the detailed design, procurement,
and installation of the shore tower at Shellman Bluff and the installation and
check-out of all communication equipment.

3.1 Recording Setup

Figure 3 is a diagram of the recording setup. The signals from the three
highest antennas are amplified at the antennas (to overcome the high loss of the
long transmission lines), and sent to a down-converter where the signal is mixed
down to a 70 MHz IF and sent to the quadruple-diversity combiner. The recording
apparatus· coupled off a sample of each of these fO\lr 70-MHz signals and
converted the power level of the individual IF signals to dc voltages to drive
the strip-chart recorder. A time code was placed on one of the six channels of
the recorder to mark hours and days. The levels of signal for each radio
channel were calibrated using a signal generator to simulate the inputs.

3.2 Strip Charts

Data were recorded on the strip chart from March through May and again for
2 weeks in July. During this period, 1718 hours of data were recorded. The
chart was run at 7.5 cm (3 in.) per hour so that a day's data was 1.8 m (6 ft),
which provided excellent resolution for the fading data. Figures 8 through 12
show examples of the strip-chart records.

4. DATA ANALYSIS

4.1 Signal Level Calibrations

To determine the strength of the received signal level as a function of

strip-chart trace deflection, it is necessary to inject known levels of rf power
at some point into the receiver system to simulate. actual received power and to
note the resulting chart-trace deflection. The ideal location to inject this
calibrating signal is at the input to the first device in the receiver chain
that provides signal gain. This allows proper account to be taken of the fixed
gains and losses in the system which should be quite constant with time.
Unfortunately, this was not possible with the current measurements. The great
height of the tower at the land end made it advantageous to place rf
preamplifiers near the upper antennas to send an amplified signal down the

3
 CHARLESTON TACTS
1\'30' 10' 30' 10' "'30' "'20'

U'30' J..--------l:...---------

U' J-.......- - - - - SaYaDnah.

a.or,la I

l ....
"

.:,' t e • •'
"
:
. . , . , . . , . , .• • • • 't

"
•••• '1 •
, ::.• ~." ,I
, ........J "
I. ••••• l. •

30' 30'1--------f-l~--::_--';.:',~_++_-__1r_-_+--_+-----_r,:..;·.:.1
,,
;:..-------t----j
:,;,.:;;':rItO:::...
,.l
,, ·:
,,

,, I•
I

I
I I
I
II I
I
I
•
......ICII. IIIW .~'---ll..---Itl.-_-..JJ.L-_ _....L._--l.-_-...J;---~;::-.
M
r
••
I
.
--~.~:----;;--d.

Figure 1. Plan view of the Shellman-Bluff-to-Ocean-Tower radio link.

4
o 10 20 :J). 40 50 60 70 80
DISTANCE (kilometers)
Link Profile From Ocean Tower To Shellman Bluff

Figure 2. Link diversity configuration.

tower. Calibrations were made using the measured parameters of gain and loss
from the antenna systems. This lead to some inaccuracy in the calibrations.
The inaccuracy in calibration, the narrow range of deflection, and the non-
linearity of the calibrations makes detailed quantitative analysis of the time
distribution of signal level impossible. There is a further difficulty with
these calibrations, namely they were done only once at the start of the
recording period. Experience has shown that some drift in gain through the rf
chain will inevitably occur, and the calibration should be repeated periodically
to correct for this effect. This was not possible for this test, but the drift
over the period of measurement, which would be reflected in changes in the
normal signal levels, did not seem to be sufficient to invalidate the analysis
that was made of the data.

4.2 Propagation Regimes

During the course of the measurements, (which extended from early March
intermittently until late July) at least five clearly distinct propagation
regimes could be identified.
The first could be called the "normal line-of-sight" periods characterized
by predicted median signal levels on the upper three antennas and a depressed
level on the bottom antenna. The signals on the elevated antennas were observed
occasionally to scintillate or to experience a few dB of random-appearing
variation and, at other times, to have periods of deep phase- interference
fading. This propagation mode would be characteristic of a path whose
atmospheric conditions varied from well mixed to horizontally stratified.
A second regime is characterized by a strongly elevated signal on the
bottom antenna and high median signal levels on the upper three antennas,
usually with slow, deep fades. The deep fades on the signals on the upper three
antennas during these periods were generally uncorrelated and a visual review
of the chart showed no instance of deep fades occurring simultaneously on the
upper three antennas. These periods are probably the result of moderate surface
or low-level ducting which "guides" the signal to the lowest antenna and
provides a source of multipath signals to cause the phase-interference fading
on the upper three antennas.
A third regime is characterized by strong signals on the bottom antenna
and severely depressed median signals on the upper three antennas. Signals on
the upper antennas during these periods are also subject to rapid, deep fading,
although, again, the fades are, not highly correlated. These periods are
probably the result of strong surface or- low-level ducting that severely
attenuates the signals at the upper three antennas and enhances the signal to
the lower antenna.
A fourth regime is characterized by low signals on the top pair of
antennas and on the bottom antenna but with a nearly normal line-of-sight level
on the mid-level antenna. Such a pattern of signal levels could occur if an
elevated duct were to attenuate the signals on the upper antennas while the lack
of a lower level duct would allow the bottom ~ntenna to receive the "normal"
low-level diffraction signal.
A fifth regime is characterized by depressed signal levels on all
antennas. Since this effect is rare, its cause may be one that occurs as a
result of unusual circumstances. One possible scenario would involve a wedge-
shaped, moist air mass that was being over-ridden by a much dryer layer so that
signals from the ocean tower would be trapped and energy reaching any of the

6
 lNA ~ 311 m
1
lNA ;: lDw Nol..
lNA II:US m \
RF Prwampllfl_ Strip Chart
Recorder

IIIl '
'-I
229 m Signal Log
Generator Unear
Detectors
• 0ulI00ll
Time Code
Generator

I
49 m
1
CalIbratIon
.....
....... ';;)t r
I

.....
Down
C.....,.
Conyert.,.
---

70 "Hz IF
Coupl~
I I I Sampling

/ _
I

.j
-
Quadruple
Eq... Go'n
Comlblner
CombIned
70 MHz
IF
lI
.,.....
Demod-I,
i

Figure 3. Link received signal level recording setup.

four shore-based antennas would be severely attenuated. Since this cause of
the low-signal condition would be a fairly rapid, dynamic change in weather
conditions, such a state of affairs would not be expected to last very long.
The brevity of the periods during which RSL's in regime 5 were observed
(Figure 4) supports this thesis. This is fortunate because such conditions,
resulting in decreased signal levels on all diversity branches, seriously
compromise overall link reliability.
Examples of each of these conditions are shown in Figures 4 through 8 in
Section 5.3.

4.3 Analyzing the Strip Charts

Since the strip charts were automatically marked with hour and day
signals, the first step in the analysis was to time and date the charts.
Periods with no data were noted and during periods of good data, dates and times
were written on the time channel. The smallest time division of analysis was
one clock hour, since all occurrences of interest were this period or longer
and since more rapid changes were considered to be part of one of the
propagation regimes. Each hour of valid data was categorized into one of the
five defined regimes and the results are presented by month so that seasonal
effects can be observed.

5. PRESENTATION

5.1 Comparison of Predicted and Measured Signal Level Data

The single most significant indicator of the quality of the installation

of a microwave link is the comparison of the expected received signal level with
the observed signal level. Although this comparison tells nothing about the
precision with which the input traffic signal is reproduced at the output, it
does ensure that the basic structures of the link have been put into place
correctly. Table 1 shows this information for the Shellman Bluff - Ocean Tower
link. Note that this prediction is for clear line-of-sight propagation and thus
is valid only for those periods when each antenna is in that mode. The strip
chart sample in Figure 8 shows that the two highest antennas operate at or near
this level, -46 dBm, when their signals are constant, indicating normal 1ine-
of-sight conditions. For the middle antenna, the signal level appears to be
somewhat lower than the -46 dBm predicted, but this may be a result of the
calibration uncertainty or of the drift in system gain in this receiver chain.
At times, the highest level observed for this antenna, indicating line-of-sight
propagation conditions, is above the -50 dBm level while other periods show a
signal between -55 and -60 dBm. For the bottom antenna, the highest RSL
observed exceeds the top calibration at -60 dBm, while the lowest level is below
receiver noise level.
The signal levels on the highest pair of antennas on this link correlate
well with the expected value, indicating that the antennas are properly oriented
and that the installation is sound. The signal on the mid-level antenna appears
to be somewhat below that expected during 1ine-of-sight conditions for this
antenna. The level shows fairly large changes from time to time, so there is
no· definite indication of antenna misalignment. The signal on the bottom
antenna behaves as expected, changing from a level typical of a high-loss
diffraction mode of propagation to a level typical of a much lower loss ducting
mode.

8
 Table 1. Summary of Predicted Link Performance

Shellman Bluff Ocean Tower

Frequency 2.3 GHz

Path length 76.83 km (47.3 miles)
Free space loss 137.3 dB
Transmission line length 361 m 49 m
Transmission line loss 14.2 dB (-10 db RF preamp gain) 1.92 dB
Combiner losses 3 dB 2 dB
Other losses 1 dB 1 dB
Total fixed loss 8.2 dB 4.92 dB
Total loss 150.5 dB
Antenna Diameter 3.05 m 2.45 m
Antenna Gain 34.7 dB 32.8 dB
Total Antenna Gain 67.5 dB
Net path loss 82.9 dB
Transmitter output power 37 dBm
Received signal level -45.9 dBm
Threshold of performance -86 dBm
Thermal fade margin 40.1 dB
Interference fade margin 65 dB
Dispersive fade margin 45 dB
Effective fade margin 38.5 dB

Table 2. Signal Level Regimes by Fraction of Time

Regime 1 2 3 4 5
Month \
March .619 .360 .011 .003 .007
April .312 .668 .004 .016 0
May .394 .561 0 .046 0
July .823 .167 .010

All Data .505 .470 .006 .016 .003

5.2 Long Term Signal Level Effects

The observation of the signal-level performance during the spring and

summer allows us to anticipate that the link will perform overall in such a way
as to support the link traffic in an acceptable manner. Table 2 shows the
fraction of the measurement period that the signal was in each of the five
regimes. The most surprising aspect of this table is the large fraction of time
that the lowest antenna received what approximates a 1ine-of-sight signal level.
Overall, for about half of the time (0.470), this antenna could have carried the
traffic by itself. A heartening aspect of the table is the low f~action of time
(0.003) during which all signals were severely depressed. Even during this
time, all of the signals were not in deep fades simultaneously so the quadruple

9
diversity combining could have allowed the link to support traffic for a large
part of these periods.
It is interesting to see the pattern suggested by the difference in
periods during which low-level ducting (that is, regime 2) was observed in the
various months. March showed about twice as much normal line-of-sight
propagation as ducting while April showed twice as much ducting as normal line
of sight. May also showed a preponderance of ducting but not so large as April,
while July showed a smaller fraction of ducting periods than was observed in
March. Apparently, the late spring - early summer period has much calmer winds
or much longer periods of calm, leading to more and longer periods of
well-developed atmospheric stratification than the earlier and later seasons.
Figures 4 through 8 show the changes of signal regime by the hour for the
months of March, April, May, and July. The lowest trace is regime 1 and the
highest is regime 5. Note how persistent the high signals are on the bottom
antenna during the March-to-May time period--sometimes for days at a time, as
shown by the appearance of the trace for regime 2 (the next one up from the
bottom). These periods are much shorter during July.

5.3 Anecdotal Analysis

In this section, we will present examples of the various types of

propagation described in Section 5.1. Typical chart samples are shown that are
not unique but rather attempt to give an idea of the changeable conditions on
the link.
Figure 8 shows an example of the most quiescent conditions on the link.
All signals remain within a very narrow range of levels over a period of several
hours. This is not a common occurrence, and,indicates that ideal line-of-sight
conditions exist. This signal level structure would indicate that no horizontal
stratification of any intensity is present, and in fact, windy conditions would
be expected to produce this effect. This period allows the normal levels of
signal to be estimated. The signals from the upper two antennas are in the
range of -46 dBm, the signal from the mid-level antenna is about 10 dB lower,
and the signal frgm_ the bottom antenna is below -90 dBm.
The data sample in Figure 9 shows examples of regimes 1 and 2 and the
rapidity of the transition between them. At 1200 on 4 April, the bottom antenna
was operating in the high-loss diffraction mode and the upper antenna signals
were fading moderately. Shortly before 1300, apparently a low-level layer
formed, which increased the signal on the bottom antenna by more than 30 dB.
Almost immediately, the signals on the top two antennas--and slightly later the
signal on the mid-level antenna--began to exhibit the deep, slow fading caused
by the arrival at the antenna of reflected and refracted secondary signals;
th~se signals changed over periods of tens of minutes from being in phase
conjunction to being in phase opposition with the main signal, leading to the
30-to-40-dB sharp bottom fades. The deep fades on the upper two antennas are
almost always independent in time, the one exception in this figure occurring
about 1545, when both signals decreased together. In this particular case, the
level of signal on either the bottom or the mid-level antenna was sufficient to
maintain service.
Figure 10 shows a condition that occurred fairly rare ly (0.006 of the
time) but could have reduced the availability of the link if the bottom antenna

10
 March 1989
rn
0,)
Tap Low
IotId La"
Bot Lo.
I'

II
S
• ~ Tap Low
$00... T\ IIId Hlp
~ Bot Lo.
W
~

~
Top Low
t-" Mid Low
t-" 80t Hlp
o

I I
II II
No Dolo i
___ " 1 , I , I I , , I I I ! I .L ,_~_l.--_L__ 1. __ I L.. I _1 -L._ ..J" ----!. -J- _.__ -L-. __ J ~_,.-......1 __ L __ --!.. __ 1 ..L

5 10 15 20 25 30
Day of the Month
Figure 40 Propagation regimes by time, March 1989.
 April 1989
Top Low
I
rn
Q.)
=l=
S
I I
.~ Top Low
Q.o Wid Hlih
Bot Low
Q.)
~
I

t-'
N
~
0
"1""'"4
r:~ I
Bot HI;h
I
~

~
==1111-
t\j II I III.
0..

~
e I II llld ffip
'OP HIgh
Bot Low
I IIII II II
No Data
I I I I I

5 10 15 20 25 30
Day of the Month
Figure 5. Propagation regimes by time, April 1989.
 May 1989
rl'l Top Low
\I.J llid Low
C,) Bot Low

S
"I"""'l
~rI
'e(.lJ
C,)
P:::
Top Lo.
t.lld Hlll~
Bot Lo.
II I
Top Low

t-' ~ Mid Low

Bot HllJh
v.> 0 -,
"I"""'l
+oJ
CO
~ rI
\ooI.AJ
CO
Top Med
Il1d t.ted
Bot HllJh
1111 I 111111111.1
0..
o
~
l'.
.....
Top Hilih
Wid HllIh
Bot Lo.
1111 1111111111 I
No Da\Q

, 1 , , I , , I I I
•
, , I , I I I I • I , I ' I I ' , , I I 1

5 10 15 20 25 30
Day of the Month
Figure 6. Propagation regimes by time, May 1989.
 July 1989
I
rn
CI.)
Top Lo"
IolId Lo"
lot Lo.

S
.bD Top Lo.
NIcI .....
Bot Lo.
CI.)
~

....
.j:'-
~

0
......
Top~
I.Id ~
Bot 11th
I
~
CO
Q1)
CO ==
Top .....

I II I I III
0..
0~
~
Top HIgh
YIcI HIIh
Bot Lo. IIIIIIIII
ND- DatQ
I I I I
----.L...

5 10 15 20 25 30
Day of the Month
Figure 7. Propagation regimes by time, July 1989.
 ,'",I +:'
,. I, I--i-""'
-.-"
Received level, -dBm

\i:''I
, !
. __ .,
I'
I
,---r.',.-,
1 "4 i- I~ 60
'1 .'1 1I t -
. i; -+-,
I
- • ", i ' 80 BoUom

:'j"jLT, ;l~ lIL.

~-+,l,·4l rl't': +'1':-
"I' 'I
r'
"t.. --1'1+"
:-+.;'," . --','1' ' I.~
I " ' 1 I" ' "
' t--
' ! "'\"f'
i "
t'
·1
-.1 ..,
'!,
T

i - ,-+-", " ,- t
t + 'f-+-',' ,
'",
h·- 'T ,
+, , ..,,', 11.. , : ! .' :"; : 1 : i 100
Anfenna

t I, I i:ii I! I 1 ' I ' r-~t- t+h,l'-:"I:++'

Received level, -dBm
--,i-i--i.. IT
. to. __ -~
+~r f: r
, ,....,j.'··t-1-. I - -
i-+-t-l ,++.+-+
-++I+'i,1
't" .. j 1... ,W:++++++-++ f ~u~.t: !
I
-------------

-f..-. - I 1,---

.--" - .----:---+-
T~+i-
m
Hi fl+"-_:j,-I
'T
-+
i
60
80
100
Mid-level
Antenna

•.•.---.------+----+............... I I I I I --t----.-~- "-t ~~-t- --t ---+----+----t---... ---t... -. --t-- -+--.--t--1- ......

t-' Received level, -dBm

VI ...i.-..;.._ -----_.-.-
i±±±H I-H-l--;,:
~ ,-
40
60
Top-Most

t+f:~:::ltt+
Antenna
.~:+ r- 80
._-~._-~ - .1---
-+-
-+ + t- , ' ~=t Hr·!
1 ,. ---t--t--t ~............- ........ -+- ~ ..---...-..-.- ..

------j------,--.-....,..---,-.--,-....,-,TI""....,.,-"",,--,-,-""+r""-,-,-TT,..,-,..,-,,--,,--..........,+-r-.-------r------ ...... _. ,,-

I Blank

1 "!;;I: .1 i!I i , ~'--'--'-~..-'--'-~,.-L,;

48:11 /17:#11 b'-:H 05':N 6'N a: OZ'/N1

i

~.
! !
, ,
Time Marker
I Channel

I
"I 1-----1

Figure 8. An example of quiescent propagation conditions, regime 1.

 .... -+---4 ---+-+-+--+- .... -....... -+---t -+-+--+--+--+- .-+--+--t---t--+-+--+--+--+--+-+~---+-+-.'-+--f-.-t--.-".t--+_.+-
I 1 I I I I I I I I I I I I I 1 1 1 1 I 1 1--.-+-+-.·-.... · +.-+-----t-....--+---+-.--+ .•
Received .l~v~l! -dBm

.~~.
60

1JMmf- 80 Bottom
100 Antenna

Received level. -dBm

it+-
r ~ .- ~ .~ , , , :.' . . . ....---- ------ _.
~ll_
. " 60 Wid-level
+-- '''' • i.
';" _.. . 80 Antenna
+-_ ",; :::: :',i 100
r-t-.-+---.--.+--+- :;, ", ." i!;~

f-++i±±tH-t+'" ---.-+--+--.. Received level. -dBm

·tJf
. ' - + -

tHl±t$H"" , " , - , - _ . i ii'::;:":,:'::''!, ....••••.. ~

~--HYi ;'-l-+~'+ ::j: . M I l M · " - · ·
~ ;·ttT=l~~rtit-,tfi
;".H
,'~::ii, -1 : Next Highest

A"'."",
0\ &l'CMl-" HSI ·ON

~t~:~j ~ ~~~=.e<'I.~'.'"40
-d8m
Top-Wost
60
Antenna
80

-J--
,~-+-l--l
;-f---;---":'_.L--i-
:"-+--';'
.:::-:-:;·tti++++--H -.l:rrll'
:..+-l+;:;!·
1.".'1'
, .' "
.,.;':
-'-, t ,1" -.... , ..
Blank

-I=T}li~~~r~~~~ji~:itlfr.:rf: : ···lll·:J:.I:L:;~:r~··
··"·+-!i 1·t1'"
_ ~o 1-t.'4() *// 198~~:N 12:to1()
-.-.--.---.'
I II I II I I II II III
I I I I'
I I I I I Illllmlml·III·111+-I:ll I i .
'I •
I! ~ '!": . ; -,f-
r- ~
.
•
I ,
I Time Worker

: I;II1\'1I Channel
I I; I

1., i ir I!r,I I I
r i I jlJ i 1 II
r-- ---. -r·.,.-.....,..-r·r--r-~·..,..--..r-''''--' t-----r--+--r--,.-l .-,-. ....... r-r-f ...... ~ -+--+-t" ... - . -.... _..... -f--t--+ .-t~-... -+- _·1 ~-+-..............- t--r--+--1 - _ .... --t--+.1<--;----fo ~"""""""-,

Figure 9. An example of the transition from propagation regime 1 to regime 2.

 I I I I I I I I I I 1--"'-' ,.,.--+.-t- ......... 1-' • • f I I I t I I fl. I I I I I I • I I f I I I I t I I I I I I I I I I I I. I I I I I I I f I
Received Level. -d8m
111 60
.. ~
~ ! ' :~ 80 8oUom
·······-·······Irl ··fTflT r'I. i . 100
. Antenna

I, Ii
"1 1 i '

. . I' Received Lev.l. -d8m

R+++++~ ++4 I I I I I I I I II I I I I ++++-+-1+
A+++~ 4I. 4~ 4+I 4 ~ i I I ++" +I I I I I I I I I I I I I I I ~ I I I I 1 I I I I 11 I I I
t

_. .. _..
, '"--::E1tttt1t
±;I;;;il'..t...1""t..
·'1
" ,
~l""H-;-I.;;L+++++I·+-lI++++++++++++
. :. .,..,. 'u''-t"'t"'!'" . ,,...:... ...• - ... - C. ...
+! :
'1"'1'
1-' ;.. ·_·1·
60
80
Mid-Level
Antenna
. '., ~ 100
,::-f
Received Level. -d8m

I'11111111·,I:!:"III·,IIIIII~
40
Next Highest

'", ~ ~:i
60 Antenna
80
: ,' "" . ._,_
_- .... I II·hR-" ...... --
::vceived Level. -dBm

1111111""II"li:~:, ,I'I'1 ':;~;:;":.~

r-'
'-J

40 Top-Most
60
~
Antenna
80
at
- _'

- + -I-~
__
j
" " . ~-~ ..~ ~~-,

~
;1111~liii
-

.~
.. ~.. f-+-+ ~ . .. .... ,.. . .

~"
~

f
.-.

':" '::..r-
.+~.-j.-
1.-to1 ·!'"1
...,.,
'T:
810nk

Os:' 110 /94,,,.;,. /P89

: ,n) tN.. b·..."
mJl!
07:dO OZ-.JI1 .
..........

·-tlit ill' ,=nt,t"lirM Time Marker

"I'I'I~I'
Channel
." . I I
ii
.'.: .... " i .

Figure 10. Regime 3, bottom antenna showing the highest signal.

had not been installed. From about 0200 until 0400 on 19 April, the signals on
the three highest antennas were significantly reduced below normal, while the
bottom antenna's signal was quite high. Even though the signals on the upper
three antenna's could have maintained service during this particular incident
(from Table 1, note that the performance threshold is -86 dBm) , this is a good
indication of the efficacy of the diversity scheme.
Figure 11 illustrates the appearance of signals during regime 4 when
signals were low on the top two antennas and on the bottom one but the mid-level
antenna was maintaining a more normal level. From Table 2, this condition is
observed for 0.018 of the time, so the diversity system is significantly
improving link availability.
The condition shown in Figure 12 is the one that will cause the most
serious decrease in link availability. During this and similar periods, the
bottom antenna received very low "normal" signal while all three upper antennas
simultaneously operated at seriously reduced levels. There is apparently no
diversity arrangement that could provide protection during this sort of fade
event. The link would not have been out of service for this entire period
because the combiner would have added the signals for the best output on a
continuing basis. It does not appear from a careful inspection of the chart
that all signals fell below -86 dBm simultaneously for more than a small
fraction of the entire period.

6. CONCLUSIONS AND RECOMMENDATIONS

Even though periods of low signal on all diversity branches were observed,
these periods were short and the total time during whidh the link would have
exhibited error rates in excess of 1 in 10 6 would be very short. This level of
performance fully justifies the diversity design.
The link was installed properly, as far as can be determined from these
measurements, and should continue to perform to the design level. The signal
level on the mid-level antenna should be checked during the next preventive
maintenance routine to determine if possible the reason for its apparent
depression.

7. ACKNOWLEDGMENTS

The author wishes to acknowledge the support of Mr. Marvin Ranta of the
Naval Air Test Center for his support of the project and Mr. John Meirose of
NATC for assisting with the data collection effort. Thanks are also due
Mr. Lauren Pratt of ITS for his effort in assembling the test equipment and
setting up the measurements in the field.

18
 .• -.-.~_.• - • • - • . • -.

-::rl
I -.-.-~- • . • --t--+ _ _ -+-+--+ I I I I I I _~-~ -. -+-_.~_._ .
Received level, -dBm
·.··\·1·
..... 60
80 Botfom
F'f1"'R'~_,.,.
r+: -. - .
rrJ··j···· 1-
·rrr 1 :·111 I"lI 100 Antenna

1. 1 l :or n:::J Ol ::r:r:u::r:r::w:J:LJ :orU::::l

il,::rOl
I I Received level, -dBm
j 11"~-r:-'
; i :::JI i . , I

. :; 11"'; :i:·..i:·:·t _..:. -.._.,.- .-

I

r-- -·i . . i I I: i I I I I I 1 I I I I I I I I I I I 1 I I I I I I I I I .. 1 I I ITT I I I I I I 111 I I rI I I I I I I I I I I fTTTI-- Mid-level

60

.
: ' t···, .,:·r:l.~·
.J I I
:1 :·I--~­ 80 Antenna
•.• '1
.-! rrtt:::::::t
I I I ' , • " , , , , , , , , , , I ' , , '- .
100

_+.,i .. I
~-t
--·-__ 1
i
+.-t
I . I I , ,
I

, , 1 , , , i ,
I

i ,
I' I I

I .M.......N·'
I
I,
I

I , iii i
I

i.'
I I

, ..,
I I

I ,
I

, ,.
I

,
I

,. ,
I

,
I.
I,,, i , , , .. , ,., , i 'i
I I

'i
I I

iii
I I
iii
I I

i
Received level, -dBm
--

.--, ..J.._
~~
.+ .... -,---t
., yo- •• _.;-+- 40 Next Highest

._--
._-.- ...-
.---,--
.. -
. 60
80
Antenna

_.......• -.. - .-..,--~

....

._
IMSa ..

.~ ~ ,~_.~~~~.~~I~~~~"~L~~'~~i~~~~I~~i~~~~~~~~~I~~~~~~~~i~~i~I~~~=R=e_c_e_iV::iod
\0

__..__...
__....__ . _.~
__
60
80 level, -dBm
Top-Most
Antenna

-._- -- .__.

;-r-H . ,. i ++ ,. , 'L+++_'
·rtf·ff13'
-'- -T-"-- ;I"j'
.....- - .. , .
.-. -:++~
.... - ·-t--t--l
.LIT·::1 :1 T··'~·
.. -•• -;-··....···t ._. .._.

. 'j ··r::·_:::·· 1, ' 1 1 lor

. -f-""··· '·1 '1-·t"'
Blank
-~:....: - -'-'1 •
+ !!
:r I'
II' I JII,! .','
,j I
l'ltl:t,..+-r:1
' :+f+i-rn:::f
~.,
Ii 4r r 2'."14 72:116 I 7/ .-011 -",".~1. ... '. /#'4" /4.-""
.i' II
Time Marker
Channel

I I I I I I • I I I I I I I I I I I I I I I I I ... I-T 1 __ ,~,.-...... "'--1"-1 - - t - - r + - - - 1 """T"""""""O"' •• _

Figure 11. Regime 4, middle antenna showing the highest signal.

 1 Recei"ed Le"el. -dBm

60
80 Bottom
.... r..;",....
''ht'i--.- ... _···.. ··--;,-;",'b -;--~_. -: .. !' d.·~ -_.. ..... __ n...... ;- .._.. _.po. --'_ . ~~-~- -- _:-:-~-_._.-
Antenna
100

---_._----_.--_.- - ,-. __ . _.
Recei"ed Le"el. -dBm
---------.---t-- -- ----------t-. .- t--- ------

60 Wid-Le"el
.. , ' ............. ' 0 ...... -~·-~W" ""'~P......, ••.. ~=.:.=
__
80 Antenna
100

Recei"ed Le"el. -dBm

------'--------:...--------:----:-::-~~~-:-::----:-~-:----:---
~---_._- . -- ----'-
I __

==-.r+·~--..,V0'"f~=~~t"
40 Next Highes-
60 Antenno
80
r-..>
o i I
Recei"ed Le"e'. -dBm

40 Top-Wast
60
Antenno
:-:1 80

Bionic

--- -'------------- . II - 47lifiirc~---I'e--i­

.J1·~ ._---- - . .!l,:~t - '5:':~----r"'"
I - I
,. jq~ __._~~_
! i I I
i ' I,
'.
-
I
--
-
(3 j
•
'JIJ: I t!''''~

!;:~ I 1.---
I Time Marker
_- Channel

I
i I ---_.
1 I"; ; •• ;

Figure 12. Regime 5, all antennas showing severely depressed levels.

 8. REFERENCES

Barnett, W. P. (1979), Mu1tipath fading effects on digital radio, IEEE Trans.

Commun. COM-27, No. 12, December.

Brennan, D. G. (1959), Linear diversity combining techniques, Proc. IRE ~,

No.6, pp. 1075-1102.

Crane, R. K. (1980), Prediction of attenuation by rain, IEEE Trans. Commun.

COM-28, No.9, pp. 1717-1733.

Dougherty, H. T. (1969), A survey of microwave fading mechanisms, remedies and

applications, ESSA Report ERL 69-WPL4. (NTIS Order No. COM-17-50288).

Hause, L. G. (1981), Selective fading on a long 8 GHz 1ine-of-sight path in

Europe, NTIA Report 81-84, September, 101 pp.

Kirk, K. W., and J. L. Osterholz (1976), DCS digital transmission system

performance, Tech. Report 12 - 76, Defense Communication Engineering Center,
1860 Wieh1e Ave., Reston, VA 22025.

Vigants, A. (1975), Space diversity engineering, Bell System Technical

Journal 54, No.1, January.

Wortendyke, D. R., A. P. Barsis, and R. R. Larsen (1979), Signal level

distributions and fade event analyses for a 5 GHz microwave link across
the English Channel, NTIA Report 79-30.

Zebrovitz, S. (1975), The use of multiple diversity to minimize the effects of

ducting on a long microwave path, Proc. IEEE National Te1ecommun. Conf.,
October 23-28.

21
 APPENDIX: LONG-TERM LINK RELIABILITY PREDICTIONS

During the design phase of this system, ITS assisted NATC in assessing
various alternatives for this communication link. Two of these analyses are
discussed here to provide background for the data collection and analysis. The
first analysis considered the expected performance of a link over this path
considering only the effects of rain attenuation and multipath fading. The
second analysis considered the path as one subject to diffraction fading arising
from the effects of sub-refractive conditions, as well as the effects of surface
reflections.

A.l The Rain Attenuation and Multipath Analysis

This analysis utilized a set of programs prepared at ITS [Hause, 1986]

that have as inputs the geographic, physical, and climatological information
about a proposed radio link and, based on a particular equipment selection,
~roduce an expected distribution of single-receiver basic transmission loss and
a receiver carrier-to-noise-ratio distribution. The path geometry and
meteorological information input and the calculated outputs are shown in
Table A-I. Table A-2 shows the inputs of the remaining meteorological
parameters and the calculated basic transmission loss. Figure A-I shows the
profile on which these calculations were based. Note that the mid-level and
bottom antennas are not considered in this analysis. Figure A-2 shows the
distribution of transmission loss due to multipath effects based on the methods
given in the International Radio Consultative Committee Recommendations [CCIR,
1982]. In order to develop a distribution of carrier-to-noise ratio, the
effects of diversity gain and the effects of rain attenuation (which produce the
same attenuation to the signal on all diversity branches of this link) must be
considered. The method of calculating diversity improvement is from CCIR
(1982), and rain attenuation is calculated for various fractions of the year by
a method developed at ITS [Dutton, 1984]. Table A-3 gives the input parameters
and the rain attenuation for various fractions of the year, and Table A-4 gives
the output of the analysis. The calculated error-free-second availability is
compared with the DCA requirement [Kirk and Osterho1z, 1976], and although that
requirement is not met, the link meets the design reliability of 0.99995
established by the Navy. Figure A-3 shows the carrier-to-noise distribution for
the link. The results of these calculations indicate that the link will support
the traffic with the desired reliability.

A.2 The Obstruction Fading Analysis

The effect of subrefractive atmospheric conditions on the performance of

microwave links has been known for some time but recently, work has been
undertaken to quantify these effects [Schiavone, 1981; Vigants, 1981]. These
algorithms have been coded at ITS for the U.S. Army Information Command in a
program called POMMP (Performance of Mixed-Mode Microwave Paths). The
parameters necessary to operate the programs are given in Table A- 7. The
analysis takes into account measured information on the fraction of time that
sub- and superrefractive conditions occur in a particular geographic area. The
data needed to perform such an analysis are obtained from radio sonde readings,
which measure air temperature, pressure, and relative humidity at intervals as
the sonde is carried aloft on a balloon. These measurements are taken at fixed
locations twice a day and, if the radio path to be analyzed is near such a

23
weather station, a time distribution of refractive gradient in the lower 500 m
of atmosphere can be obtained. An example of this distribution for the Georgia
coastal area taken from Samson (1975) is shown as Figure A-4. A technique
developed by the author and Mr. Hause [Hause and Farrow, 1985] permits the
reduction of these curves to a bi-modal distribution similar in effect to the
method given by Schiavone (1981) to facilitate the analysis. Table A-8 gives
the input parameters for the calculation as well as the resulting refractivity
gradient distribution; Figure A-5 shows the calculated bi-modal distribution.
As the refractive gradient changes from the normal level of -57 N-units per km
of altitude (which corresponds to an equivalent Earth's radius factor, k, of
4/3) to a N-units per km (k - 1) and on to positive gradients of +157 N-units
per km (k - 0.5) and more, the Earth can be visualized as "bulging" up toward,
and finally into, the radio path. In the other direction, the k-factor
increases without bound giving a "flat Earth", and for larger negative values,
a concave Earth with a negative radius. The effect of Earth bulge is
illustrated in the sequence in Figures A-6 through A-ll. The basic
transmission loss over a pat~ depends on the clearance of the beam over the
terrain, and the loss increases very rapidly as the clearance goes from positive
to grazing incidence and then to negative clearance or obstruction. This
increase in loss is shown in Figure A-12, which is the distribution of combined
path loss, and in Table A- 7, which shows the tabulation of the various
components of loss.

A.3 Discussion

It is coincidental that both techniques for predicting the most severe

attenuation over the path arrive at similar conclusions, since quite different
methods of analysis are used. Comparison of the extreme attenuation exceeded
for 0.00001 of the year (shown in the lower right-hand corner of each graph) on
Figures A-2 and A-ll shows values of 186.5 dB and 181.5 dB respectively, with
the lower value resulting from the multipath calculation. In interpreting these
data, however, it must be understood that the multipath attenuation effects are
susceptible to improvement by space or frequency diversity while the diffraction
or "Earth-bulge" fading is not, which would indicate that link reliability will
be limited by the system gain available to overcome the diffraction fading.
It is important to bear in mind that neither of these prediction
techniques considers the contribution of the mid-level and bottom antennas to
link availability, nor does either technique suggest the utility of the extreme
diversity spacing used. As the measured data show, the bottom antenna can
provide adequate signal to carry traffic during a considerable part of the time.
The link design used in this case has evolved over a number of years from
analyses of the performance of a number of links and considerations of
atmospheric effects not addressed by the best currently available models. Even
the mixed-mode prediction technique has an inherent assumption that the vertical
gradient of refractive index is constant over the elevation through which the
signal travels, and this is known not to be true -for all periods of time and is
probably untrue for most periods, as can be inferred from the data presented
here. The standard prediction models perform well when they are used within
their range of applicability, but must be used with caution when the limits are
exceeded. Paths such as Shellman Bluff - Ocean Tower need the careful and
individual engineering approach that the Naval Air Test Center took in this
case.

24
 The link design for this path was intended to be quite conservative and
some cost was incurred by the installation of the lower two antennas and
associated radio equipment. This incremental cost was not a large fraction of
that of the total link installation, and the performance of the link is improved
to a noticeable extent by the presence of these lower antennas.

A.4 References

CCIR (1982), Recommendations and Reports of the CCIR, 1982; Volume V;

Propagation in non-ionized media, lTU, Geneva, Switzerland.

Dutton, E. J. (1984), Microwave terres.trial link rain attenuation prediction

parameter analysis, NTIA Report 84-148, April, 194 pp. (NTIS Order No.
PB 84-207984).

Hause, L. G., and J. E. Farrow (1985), Propagation predictions for marginal LOS
microwave paths, Proc. 1985 IEEE Military Commun. Conf., Boston, MA,
October 20-23, pp. 369-373.

Hause, L. G. (1986), Algorithms used in ARROWS: autodesign of radio relay

optimum wideband systems, NTIA Report 86-207, October, 138 pp. (NTIS Order
No. PB 87-l26405/AS).

Kirk, K. W., and J. L. Osterholz (1976), Des digital transmission system

performance, Tech Report 12-76, Defense Communication Engineering Center,
1860 Wieh1e Ave., Reston, VA 22025.

Samson, C. A. (1975), Refractivity gradients in the northern hemisphere,

U" S. Department of Commerce, Office of Telecommunications, OT Report
75-59.

Schiavone, J. A. (1981), Prediction of positive refractivity gradients for line-

of-sight microwave radio paths, Bell System Technical Journal 60, No.6,
Part I, July-August.

Vigants, A. (1981), Microwave radio obstruction fading, Bell System Technical

Journal 60, No.6, Part 1, July-August.

25
 Table A-i. Path Geometry Input and Calculation Output Table

May 19, 1986 Project:910;409 JEF

Path Geometry and Meteorlogical lnfor-ation
Ocean Tower R2 to Shore TOwer

For the R2 RV1 Path

Path Length (kll) 76.8268

Carrier Frequency (GHz) 2.0000000
Transmitting Primary Antenna Height Above Ground (m) 47.244
Receiving Primary Antenna Height Above Ground (.) 361.188
Mean Terrain Elevation (m) 0.003
Standard Deviation of Elevations (II) 0.0;8

Path Azimuths from True Harth:

R2 to RV1 286 ;8' 48.4"
RVl to R2 106 34' 33.4"

Path Azimuths from Magnetic North:

R2 to RV1 (d/lllls) 286 ;8' 48.4"
RVl to R2 (d/m/s) 106 34' 33.4"

Primary Antenna Values For Average Earth's Radius Factor:

Average Earth's Radius Factor 1. 41;9630
Transmitting Antenna Elevation Angle (d/fa/S) -DO' 26.7"
Receiving Antenna Elevation Angle (d/m/s) -28' 48.8"
Effective Path Length (k.) 76.8268
Minimum Ray Path Absolute Clearace Distance (km) 2.0000
Minimum Ray Path Absolute Clearance (m) 47.207
Minimum Ray Path 1st Fresnel Zone Clearance Distance (k.) 14.0000
Minimum Ray Path 1st Fresnel Zone Clearance (1st Rad) 1.3;99
Mean Atmospheric Pressure on the Ray Path (kPa) 99.43;0;17
Mean Ray Path Height Above Ground (m) 141.173

Primary Antenna Values For Extreme Subrefractive Earth's Radius Factor:

Extreme Subrefractive Earth's Radius Factor 0.6666667
Minimum Ray Path Absolute Clearance Distance (km) 20.0000
Minimum Ray Path Absolute Clearace (m) -3.94;
Minimum Ray Path 1st Fresnel Zone_Clearance Distance (kill) 20.0000
Minimum Ray Path 1st Fresnel Zone Clearance (1st Rad) -0.084

General Link Information:

Estimated Annual Worst Month Average Snow Depth at R2 (m) 0.000
Estimated Annual Worst Month Average Snow Depth at RV1 (m) 0.000
Tree growth at R2 is estimated at 0.000 II per year.
Tree growth at RY1 is estimated at 0.000 mper year.
26
Table A-2. Meteorological Parameter and Basic Transmission Loss Output Table

19 May, 1986 Project:9105409 JEF

Median Basic lranslission Loss

Ocean Tower R2 to Shore Tower

Path 1 Ocean Tower R2 to Shore Tower

Path Length (kill) 76.827
Primary Frequency (GHz) 2.0000
Average Summer Temperture (Celsius) 26.7
Mean Path Pressure (kPa) 99.4351
Absolute Humidity (gil/cu. Ill) 18.00
Oxygen Absorption (dB) 0.424
Water-Vapor Absorption (dB) 0.035
Total AtlOspheric Absorption (dB) 0.459
Free Space Basic Translission Loss (dB) 136.181
Total Path Median Basic Translission Loss (dB) 136.640

27
 19 May. 1986 Proiect No. 9105409 JEf DISTANCE (miles)
0 6 12 19 25 31
700 t= iii iii iii • • • 'I i i i i i i • •• 1 • iii Iii iii • i • i I •I i l l ••• i i i ••• I • Ii. I
--I 2300
lower Base Antenna Height lal
k • 1.42 Elevation lal Pr1aary Diversity
ke- 0.67 R2 0.00 47.24 33.53
ciOO ~ --I 1972
RYl 3.05 361.19 347.47
...................... 0.6 fresnel Zone
at 2.0000GHz Path Length- 76.83 k.
U)
L.
eu ~t 1 16 3
•
~
eu
~
eu
..§
400 t314
--
eu

-I
-I Ul
Ul ~
%
UJ
UJ :>
:> Cl
a 300 986 co
co -«
tv ~
(Xl z:
z: a
a t-t
t-t l-
t-
~
:>
UJ
200
/ .. 657 4:
:>
UJ
-I
-l UJ
LU
/ ....
100 V ~ -.:j 329
/ ....

O~
R2
• ?nz
-
!
,
.............................
I • ! ! • I ! ! I •
---
I • ! ! !
1

I
t

! ! ! I I ! I ! ! I ! ! ! I I ! ••• I ! ! ! ! I , , • ,I ! ! • ! ! !
K

! !
RYt
, I
'II --1 (J

0 10 20 30 40 50 60 70 80
DISTANCE (kilometers)

Path Profile from Ocean Tower R2 to Shore Tower

Figure A-i. Path profile for the multipath attenuation distribution calculation.
 19 May. 1986 Project No. 9105409 JEF

151
en
en
o
-l
c::
.....o
.....~ III
S
en
c::
~

....
c:..

•~ 170
en
~
CO
c:..
Q.l

.....>
QJ III
u
QJ
c::
,...,
QJ

O'l .J .5
.....c:: Fraction of Time that the Corresponding
c.n
Multipath Basic Transmission Lass is Nat Exceeded

Path Length kl 76.827

Effective Pith Length kl 76.827
PrlllrY Frequency GHz 2.000
PolarlnUon Vertical
Median Balic Transllssion Lass dB 136.640
Std a.viatlon of Terrain Elev I 0.151
Averag. Annual Teaperture Celsius 19.722
Region COlStal
Annull Occurance Factor 0.72163

Figure A-2. Multipath attenuation distribution.

29
 Table A-3. Carrier to Noise Probability Dlstribution Parameters

May 19, 1986 Project:9105409 JEF

Carrier-to-Noise Probability Distribution Paraleters

Ocean Tower R2 to Shore Tower

Path 1 ParaMters - Ocean Tower R2 to g,ore Tower

Path Length 76.827

Transmitter Antenna Type Parabol ic
Receiver Antenna Type Parabol ic
Transmitter Final AMplifier Type Constant power
Primary Carrier Frequency (GHz) 2.0000
Diversity Carrier Frequency (GHz) 2.0000
Transmitter Antenna Gain (dBi) 31.5734
Primary Receiver Antenna Gain (dBi) 33.5116
Diversity Receiver Antenna Gain (dBi) 33.5116
Site Loss on the Receiver End of the Path (dB)
Median Basic Transmission Loss Across the Path (dB) 136.6395
Rain Attenuation for PSO.01 (dB) 0.0427
Rain Attenuation for PSO.001 (dB) 0.3550
Rain Attenuation for Pa O.0001 (dB) 0.9695
Rain Attenuation for PSO.00001 (dB) 1.5048
Rain RadOM Loss Due to Rain (dB) 0.4000
Annual Multipath Fading Occurance Factor 0.7216
Diversity Type Space
Vertical Diversity Spacing (m) 13.716
CoMbiner Switching Threshold Power Ratio (dB) 3.0000
Transmitter Feeder Line Length (m) 50.2920
Translitter Feeder Line Loss per 100 meters (dB/1001) 1.8500
Transmitter Feeder Line Loss (dB) 0.9304
Translitter Diplexer Loss (dB) 1. 0000
Receiver Feeder Line Length (81) 370.3320
Receiver Feeder Line Loss per 100 leters (dB/lOOm) 1. 8500
Receiver Feeder Line Loss (dB) 6.8511
Receiver Diplexer Loss (dB) 1. 0000
Transmitter Final Alplifier Power Output (dB.) 40.0000
Transmitter Power at the Antenna Terminals (d8ID) 38.0696
Receiver Hoise Figure (dBl 7.0000
Receiver Bandwidth (Itlz ) 7.0000
Receiver Thermal Noise (dBml -97.5490
Median Received Signal Level for the Pa1h (dBml -41.3362
Median Thermal C/N at the Receiver Output (dBl 56.2128
Median C/N at the Transmitter (dB) 60.0000
Median Path C/N at the Receiver Output (dB) 54.6958

30
 Table A-4. Digital Link Performance Parameters

19 May, 1986 Project:9105409 JEF

PCM-TOM PerforMance Para.eters

Ocean Tower R2 to Shore Tower

Total Length of the Link 76.8268

Data Trans.ission Rate 6.300
Modulat ion Type QPSI(
Mfg Bit-Error-Rate Threshold (dBm) -79.00
Mfg Bit-Error-Rate at Threshold 0.0000000
Median Received C/N (dB) 54.696
Median Received Signal Level (de.) -41.336
Outage Bit-Error-Rate 0.00000100
Threshold Received tIN (dB) 15.012
Fade Margin (dB) 39.684
Desired Error-Free-Second Availability 0.99999~20
Calculated Error-Free-Second Availability 0.99999910
Is PerforMance Satisfactory? No

Nou: These values cover propagation effects, not equiplMlnt outages.

31
 19 May, 1986 Project No. 9105409 JEF

o
.....
-.J
~ 51
~

.....tn
o
Z
I
o
-.J
1
to.
.....
~

tra 3t
U
'C
~
>
..... 21
~
c..l
~
a:
~

.=
-J
\I

....o
I

I
'C
c:

..- ..-
UJ

u ••• .... 0• • 0._

Fraction of Time that the
Carrier-to-Noise is Not Exceeded

Nulb.r at Paths in the Link 1

Rain Attenuation Confidence Sand. Q 0.900
M.diln R.c.ived C/N at Shore Tower d8 54.696

Figure A-3. Link pre-detection carrier-to~noise ratio distribution.

32
 PERCENr ()Il' rIME ORO/NArE /S Nor EXCEEDED
u tUtU ,. M '5 to • 10 &0 50 40 30 20 10 52 I 5 2 I 05 01
300
•.. -_. ClJtrlU/tlll .... D"",w/lOfI 011". .....
200 0bIN....(/ HtJdio H.lrtlCII""I)' (;,Q(/I"" .....
1-.- _. . - - -' f--
In II» (J,OIIItt1'lJt/s.(/IOO'IM'., LtI)'H
~.

,
CAPE HATTERAS .....
~
...... .... .... ~ - -F""IIGry - -. .-.ItItJ,
... .....

-
_. _. - A/lfUs/ . .Noll",*,
....
0 r-~ .. ...... .......
~
...... ...
~ I I
-- ~
I
~~. r-....
- :,:,;",
.. .
~~~
.. \,., _~

.- - - _. _. -- . - - -- ~ ~ ',.'.:'.b..
,

,
. - i - i- - ._. .. _. .- ... \.-
.
'I-

-- - .. - . - ..-- .. _. . - ... - .....

'500
.... .......- . . _. . .. .. _. _.
- .. . .. - .. .... ...
_.~

.
.....

'.00
.Ql .oS.1 1 oS I 2 5 10 20 30 40 50 • 10 10 to 15 II. •• "'1 ,...
PERCENr ()Il'rIME ORDINArE IS EXCEEDED

Cape Hattera., N. Carolina

35- 16' N, 75- 33' W 3 III 111 ••• 1.

O.ta: Radio.onde. 0300 .nd lS00Z (2200.nd 1000 LST): 8/52 - 5/57

Temperature (OP): 3.nuary 53/40, 3uly 84/72

Me.n OewpoinC (OP): Not .vailable

Precipit.tion (inche.): Annu.l 54,51 Auqu.t 6.421 April 2.99

Loc.ted on .n elonq.ted ••ndy i.land in the Outer B.nk. of North Carolina.

The Gulf Str••111 of the Atlantic Oc.an p••••• 30 to 80 kill ott.hore. A
maritime climat. with warm .WIIlm.r. and cool winter.~ R.latively windy.

Figure A-4. Measured refractivity gradient distribution.

33
 Table A-5. Bi-Hodal Refractivity Gradient Distribution Parameters

19 May, 1986 Project:9105409 JEF

Refractivity Gradient Distribution Paraleters

Ocean Tower R2 to Shore Tower

Units Value

Annual Mean Refractivity Gradient n-units/km -52.00000

Layer Thickness
Region
• 100.00000
Coastal
Probability of a Telp Inversion 0.30000
Air Horizontal Homogeneity 1. 00000
Distto a Large Body of Water 0.00000
Wind Direction Factor 0.20000
Average Telperature Celsius 19.72222
Mean Annual Precipitation llllI 1400.00000
Surface Moisture Index 23.43889
Water Proxi.ity Factor 1. 00000
Water Vapor Capacity 0.47195
Surface Moisture Factor 0.21404
Mixed Regime Std Deviation n-units/h 15.00000
Stratified Regime Std Deviation n-units/kll 91.42112

Refractivity Gradient and Earth's Radius Factor Distributions:

Probabi Iity, q Refractivity Earth's Radius

That the H' Value Gradient, H' Factor, k
Is Exceeded (n-unit5/km)

.9 -71.13910 1.82868
.5 -50.37694 1.47254
.1 -10.45983 1. 07139
.01 64.35218 0.70926
.001 131.72485 0.54375
.0001 188.78524 0.45402
.00001 238.58537 0.39686

34
 19 Mi,. 1986 Project No. 9105409 .If
_.ILilll"" 111,"111 • 1'111.,. i ,.• I • , i I

~gi~
Coastal
. . . . .1 Air Horizontal HoIogenelt,:
~

c....
IU
1.000
Q.
Probability of Teip Inversion:
- au
~
~
.&
0.300
Mind Direction factor:
.... 0.200

-
~

•
z
...;
Average July Teiperature:
26.667 degrees Celsius
Average January Teiperature:

-
....
c::
w
IU 12.778 degrees Celsius
VI

/ r,
Surface Moisture Index:
23.439
....
>0-

-
~

~
u
Distance to a large Body of Nater:
L 0.000
i -.... Annual Mean Refractity Gradient:
52.000 n-units per k.

..... ..•. I...

-211.1 t I It I ",4 I" hllli I I I I 111'1

'.1.
I I I

~cent Ti•• II. lhit N'

I..
I .... I • I

Is fleeeded
I I
1.5
I I I ...,j
1.9

Refractity Gradient Distribution

Ocean Tower R2 to Shore Tower Path

Figure A-5. Calculated bi-modal refractivity gradient distribution.

 11 MaV. 191& Project 1105401 B DISTANCE (mi les)
550 ( ,
0.00 6.22
, 12.43
I
1•• &5
I ,
24._ 31.0.
I
31.29
I
43.5S
I I l S804
Antenn. Ha1ght AJ1.uth to Other S1te '.e-o"
Latltu'e Laneltude flev lilt Pr1....V Dlv....ltv True North .....,eUc Mile
500 L -11840
R2 31 22' II.•-N 10 33' 51.5-" 0.0 47 • 34. 286 58' 4•. 4- 286 58' ....... 00' 07.0
RYI 31 34' • • O-N .1 20' 26.0-" 3.0 . 3&1 • 347 • S06 3 .. ' 33.4- 106 34' 33.4- -21' 15.s-
450 L Path L. . .th • 76.13 k.
II • 0.5 0.& or lat Freanel Zon.
...J S47&
k· 1.4125

400 l 1 S312

350
. AV1
sue

-..
II)
300 984
L
lU +J
w +J OJ
(7\
CD OJ
e
- 250 e20 ~

Z z
a 200 e& a
H H
t- t-
~ <t
>
lU
S50 492 >
W
..J J
lU W
100 32.

50 A2
Mean Sea level --....... III ...J t64

0 0
0 to 20 30 40 50 60 70 80
DISTANCE (k i lometers)

Figure A-6. Ray path for probability of 0.5 corresponding to k=1.472.

 .. Ma" llilli Proj.ct 1105401 JIU DISTANCE (mi les)
550 0.00 •. 22 12.43 le.65 24.86 31.08 _ •. __ ._._.
I I I I I I I I • ,
i . a...
~
Ant.... Hdlht Allaath t. Ot..... su.
500
uUtl* l .... ltuM fl.v'" Pr~, Dlver.tt, True th .......Uc
T• •-eU
. .1. -It. .
R2 3a 12' II 33' 5I.5~" 0.0 47. 34 • 21& 51' 2Ii 58' .... 4· -05' .....
RYt 31 34' .a zo' 211.0·" 3.0 3Ia. 341. tOi 3..' 33.4· 101 34' 33.4· -33' 11.1·
450 Path LMetb • 71.a u 1
II • o.a. 0.&0' aet fr..... l Z.... . 1476
II· 1.0ll"
400 13t2
i
350
RU . tue

......
II)
300 M.
w
L.
...,
II ---...,
..... QJ
QJ CD

-
E

Z
250 1120
-
"-

Z
0 200 &56 0
H t-t
t- t-
~ ~
> t50 492 >
au lU
-J -J
au W
tOO 32.
Mean Sea level
50 R2 iii.

0 0
o to 20 30 .0 50 60 70 eo
DISTANCE (kilometers)

Figure A-7. Ray path for probability of 0.1 corresponding to k=1.071.

 19 Ma,. 1916 Prolect 9105409 .B DISTANCE
550 F
0.00
I
&.22
I
12.43
I
III. liS
I
- ----
I
-_I.. _- -- ._-
I
.--
I I
S8O~
4
LaUtulta LORIltude fl."W
AntllllftO . . llllt
Prl....' Dlv....U'
Allauth t. Other Slt.
1rue North .......tlc
1• •
....1.
-0"
500 L ..1 I.~O
R2 31 22' a.I-N 10 33' 59.5-" 0.0 47 • ~. 216 51' 48.4- 216 51' 48.4- -15' 01.3-
IIlI 31 34' ]G.O-N II 20' 26.0-" 3.0 361. 347. 10& ].t' 33.4- 10& ].t' 33.4- -0' 23.4-
P.th L....th - 7&.13 kII
450 L
q - 0.01 0.& 0' 1st frean.l Zan.
..1 1476
k - 0.7093

400 l i 1;'12

OYi
350 IU8

.....
II)
L 300 984
w QJ +J
co +J

-
QJ
Q) Q)

-
E 250 . 820

Z
a
z
200
. 6S6 a
H H
t- t-
« «
> 150 492 >
UJ W
-'
UJ
....J
W
100 328

::J, f'
02

] I7 0
I
10 20
I I
30 40
I I
50 60
I
10 80
DISTANCE (k i lometers)

Figure A-8 •. Ray path for probability of 0.01 corresponding to k=0.709.

 19 "-f, 1916 Proi.et 9105409 JfF DISTANCE (mi les)
550 0.00 6.22 12.43 18.65 24.116 31.08 37.29 43.51
I
f I I I I
Antenn. Helght
I I
All~th
I
to Other Slt.
I l1804

1lI1le-o"
LaUtuM Lonlltudtl fl.v lIIJ Prl"'f Olv....ltf T..ue NDrth "-gneUe An,le
500 L
R2 31 22' 2It••-N 80 33' 59.5-"
-J 1640
0.0 47 • 34 • 2116 58' 41.4- 2116 58' 48.4' -17' 56.8-
IrIl 31 34' !O.O-N 11 20' 26.0-" 3.0 361. 347 • 106 34' 33.4- 106 34' 33.4' -49' 50.2'
4!50 L P.th l ....th • 15.13 U
II • 0.001. .....I 1476
II • 0.5437

400 l i 1312

350
AYl
1148

UJ
L 300 984
w IJ +J
\0 +J
OJ
OJ

- 250 OJ
E 820 ~

Z z
a 200 656 0
H
H
t- t-
« <{
> 150 492 >
I1J lJJ
..J -.J
I1J lJJ
100 328

50 R2 164

0 0
0 10 40 50
DISTANCE (k i lometers)

Figure A-9, Ray path for probability of 0.001 corresponding to k=0.544.

 19 MaV, 1916 Project 9105409 olU DISTANCE (mi les)
550 l 0.00 6.22 S2.43 SI.65 24.86 31.01 37.29 -....._ . -. SI04
I I I I I I I I I
~
Antenna Helght All~th to Other Slte Tllke-oft
L.UtuM Lonl 1tude flev'" Prl...V Olvera ltV True .....th Magntttlc Anile
!IOO L ...J S640
R2 31 22' a .•-N eo 3i' 59.5-" 0.0 41 • 34 • 216 58' 41.4- 216 58' 41.4- -19' ,. .•-
RYS 31 34' 3O.0-N 11 20' 26.0-" 3.0 361. 347 • 106 34' 33.4- 106 34' 33.4- -54' 32.2-
450 L 'ath L.... th • 1&.13 ka
II • 0.0001 --J 1476
k • '.4540

400 l 13S2

. AY1
1
350 1141

-.
II)
L 300 984
IU .v
~ ~ QJ
o IU QJ

-e 2!50 820
-
~

Z Z
0 200 656 0
H H
t- t-
<t «
> SIlO 492 >
LLI W
J -l
llJ W
SOO l21

~[ ~, f'
R2

IL
0
I
SO
I
20
I
30
I
40
I
50 60
I
70
I
80
DISTANCE (k i lometers)
Figure A-1O. Ray path for probability of 0.0001 corresponding to k=0.454.
 19 Maw, 1916 Pro)act 9105409 JfF DISTANCE (mi les)
6.22 12.43
550 f 0.00
I I I
IB.e
I
24.86
I
31.0B
I
31.29
I
43.51
I I
lB04
~
Ant..... HIlllht Azl.uth to Otha~Site 1.e-o"
500 L
LaUtu" lon. I tude Elavllll Prl..., Dlv.... it' 1~ueNorth MIlaneUc Antle
..J 1640
R2 31 22' 2It.'·N 80 33' 59.5·" 0.0 47 • 34. 2IIIi 51' 41.4· 286 51' ....... -21' .....
RYS 31 34' 3O.0·N II 20' 26.0·" 3.0 361. 341. 106 34' 33.4· 106 34' 33 ...• -51' 19.'·
450 L
P.th len.th • 7•. 13 k.
q • 0.00001 ..J 1476
k • 0.3969

400 l i 1312

350
. AY1
IU8

U)
L 300 9B4
CU ....,
..... ....,
-I'-

CU

-
E

Z
a
250

200
820

e&
-
CU
CU

z
a
H H
t- t-
« «
> 150 492 >
W W
-I -I
W W
SOO 328

~r \t r<
R2

II
0
I
so
I
20
I
30
DISTANCE
I
40
I
50
(k i lometers)
I
60
I
10 80

Figure A-II. Ray path for probability of 0.00001 corresponding to k=0.370.

Table A-6. Dominant Mode Determination and Transmission Loss Distribution

19 Mey, 1986 PrO).cI:9105409 JEF

Ooaln.nt I1od••nd Coli In.d LOll
acun Tower R2 to $ftor. Tower

Vetl.bl.1 Ind.p.nd.nt 01 R.fr.ctlvaty Gr.dl.nt:

Pith L,ngth lkal 76.827
Prl..ry Fr,qu.ncy (1t4z I 2000.000
R2 Trentaltt.r Ant,nn. Oi...t.r (II 2.4J8
RV1 R.c.lver Ant.nn. Oi... t.r (a) '.048
Oiverllty Ant,nn. Oi,..t,r (al '.048
R2 Tr.nlaltt,r Ant,nn. Glln (d8,) H.Sn
RY1 ReC'lver Ant.nn...in (diD n.512
Oiverllt~ Ant.nn. Glln (d8i) ".512
Lin.-ol-Ilght e.'IC Tr.n,alilion Lol. (dS) U6.6J9

V.rllbl.1 Otptnd.nt Upon R.fr.ctlvlty Gr.dl.nt:

CllIU lit iyt (»),t.Oillr. ToUI Antlnn. nini_ Sltooth Eltth Tropolph.rlC
Prob.blll ty Stlic Po Int Inq Loll CI.lrenct O,IIr.ctlon Selt tlr Blllc
q Tr.n'.II,ion (dBl n,t Fr."'tl BtaIC Tren,. Tr.nt. Lou
LOll Red'al Loll (dB>
(d8) (d8)

0.9 H6.6n46 0.001" 1. 62448 152.18846

0.5 H6.""1 0.00001 1.40n2 152.18941
0.1 H6.64124 0.00648 0.96472 152.19124
0.01 14}.86'61 0.0'''1 '.OW2 152.19470
0.001 161.68209 0.096}. -0.74211 164. '7817 168.H600
0.0001 1]15. ''''~ O.lJ44' -1.46}'2 17'.1'141 178.98848
O. 00001 181.72544 0.170}2 -2.1"69 182.5"'4 184.68116

Culula.tive Antt"'" To R2 Antlft"a RYl AntlM. Coliintd Ooain."t 1'04.

Pro..-ility ntdiua Point ing \'011 Point In9 Loll StIIC Tr.nt.
Coup IiftCJ Loll (d8) (dS) .nd Pointing
" (d8) LOll
(dS)

0.' 2.'H~ 0.00110 0.00172 U6.64821 Lin. of Sight

0.5 2.)H~ 0.00000 0.00101 1)6.6J941 LIn. of Slgl'lt
0.1 2.'H~ 0.00407 0.006" U6.6-m Lin. of Sight
0.01 2."'46 0.0'''' 0.0'26' 14'.91921 (I),t.c 1. Olf F
0.001 2.'''~ 0.04798 0.1071' 16'.5'821 SlIlloth E,rth
0.0001 2.H'~ 0.0'7"" 0.1"11 172.69H2 Seooth E,rlh
0.00001 2.'''46 0.06568 0.20860 181.39n4 Saooth E.tth

42
 19 May. 1986 Project No. 9105409 &ff

131.1

141.0

150.1

lliG.1
'iO
.:!!
en
II)
o
-'
....
.c
CD
Q..

.,
.-
"C
C

iI
o
~
~

""
Path Type ~ Marginal line-of-Sight
Path Length • 76.83 k.
Carrier Frequency· 2000.00 MHz
Vertical Polarization
Median Basic Trans.ission Loss
Plus Antenna Pointing Loss • 136.64 dB
1.1 1.5 1.9 O.!I!I 1.999 0.9999 '.!I!I!l!I9
fraction of Tile the Corresponding COibined loss is Not Exceeded

long-Terl Colbined loss Distribution

Ocean Tower R2 to Shore Tower Path

Figure A-12. Mixed mode combined long-term path loss distribution.

FORM NTIA-29 U.S. DEPARTMENT OF COMMERCE
(4-80) NAT'L. TELECOMMUNICATIONS AND INFORMATION ADMINISTRATION

BIBLIOGRAPHIC DATA SHEET

1. PUBLICATION NO. 2. Gov't Accession No. 3. Recipient's Accession No.

4. TITLE AND SUBTITLE 5: Publication Date

Design and Performance of a Lpng, Over-Water Microwave February 1990
Radio Link 6. Performing Organization Code
NTIA/ITS
7. AUTHOR(S) 9. Project/Task/Work Unit No.
Joseph E. Farrow
8. PERFORMING ORGANIZATION NAME AND ADDRESS
National Telecommunications & Information Administratior
Institute for Telecommunication Sciences 10. Contract/Grant No.
325 Broadway
Boulder, CO 80303
11. Sponsoring Organization Name and Address 12. Type of Report and Period Covered
Naval Air Test Center
Commander
Patuxent River, MD 20670-5304 13.

14. SUPPLEMENTARY NOTES

15. ABSTRACT (A 200-word or less factual summary of most significant information. If document includes a significant bibliography or literature
survey, mention it here.)

This paper discusses the design and the measured performance of a 76.8-km
(47.7-mi) over-water 2.3 GHz microwave radio link. Since one of the terminals
was located on an ocean tower, the maximum antenna height at that location was
restricted to 49 m (160 ft). A 367-m (1200 ft) tower was used at the shore end
to provide ray path clearance for normal and extreme refractivity gradients.
The tower at the shore end supported antennas at 361 m (1180 ft), 345 m (1128 ft),
229 m (750 ft) , and 49 m (160 ft) to provide quadruple diversity protection.
Received signal level measurements were made over the link for several months
including some of the summer period. Propagation performance of the link during
.this period was sat~sfactory.

16. Key Words (Alphabetical order. separated by semicolons)

2 GHz propagation; long over-water path; quadruple diversity performance analysis;

refracti~e gradient analysis

17. AVAILABILITY STATEMENT 18. Security Class. (This report) 20. Number of pages

Kl UNLIMITED. Unclassified 44
19. Security Class. (This page) 21. Price:
0 FOR OFFICIAL DISTRIBUTION.
Unclassified
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I