Path Loss Channel Model for Inland River Radio...

Research ArticlePath Loss Channel Model for Inland River RadioPropagation at 14 GHz

Junyi Yu1 Wei Chen1 Kun Yang2 Changzhen Li1 Fang Li1 and Yishui Shui3

1School of Automation Wuhan University of Technology Wuhan Hubei 430070 China2Super Radio AS Oslo Norway3School of Information Engineering Wuhan University of Technology Hubei 430070 China

Correspondence should be addressed to Kun Yang kunyangntnugmailcom

Received 3 October 2016 Accepted 14 December 2016 Published 16 January 2017

Academic Editor Claudio Gennarelli

Copyright copy 2017 Junyi Yu et alThis is an open access article distributed under the Creative Commons Attribution License whichpermits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

In this paper a propagation path loss model for inland river is proposed by three improvements compared with the Round EarthLoss (REL) model for open-sea environment Specifically parameters optimization uses Okumura-Hata model in dB scale toreplace the equation transformed from the free space loss in REL model secondly diffraction loss caused by the obstacles (eglarge buildings bridges or some other facilities near the river bank) is also taken into account mixed-path methodology asanother improvement is used for Inland River (IR) model because the actual propagation environment between transmitter (TX)antenna and receiver (RX) antenna contains both land part and water part The paper presents a set of 14 GHz measurementsconducted along the Yangtze River in Wuhan According to the comparison between path loss models and experimental resultsIR model shows a good matching degree After that Root Mean Square Error (RMSE) Grey Relation Grade and Mean AbsolutePercentage Error (GRG-MAPE) PearsonCorrelationCoefficient andMeanAbsolute Percentage Error (PCC-MAPE) are employedto implement quantitative analysisThe results prove that IRmodel with consideration ofmixed path and deterministic informationis more accurate than other classic empirical propagation models for these scenarios

1 Introduction

As the third longest river of the world Yangtze River playsan important role in Chinese water transportation Ourinvestigation in Changjiang Maritime Safety Administrationshowed that the whole basin navigation mileage of YangtzeRiver was up to 80000 kilometers (including main-streamand tributaries) accounting for about 65 percent of Chinainland waterways navigation mileage In 2015 large-scaleports along the trunk line of the Yangtze River completedcargo throughput 2118 billion tons Thus it is essential toprovide high data rate and reliable communication servicebetween ship to ship and ship to shore to ensure the safety ofnavigation However the performance of present broadbandcommunication systems have not been optimized for themaritime applications due to lack of a comprehensive InlandRiver channel model which is indispensable to the systemdesign and optimization To solve the above challenge wepropose a dedicated IR model in this article

The inland river terrain profile contains both land andwater regions For the land area various measurement cam-paigns have been performed to develop the radio channelmodels (eg Okumura-Hata empirical model [1ndash4]) TheCOST-231 Hata model [5] extending frequency adaptionof Okumura-Hata is valid in frequency range from 15 to20GHzHowever some of key propagation phenomena havenot been considered by either of them including scatteringdivergence and shadowing REL model derived from Two-Ray theory is proposed in [6] for the open-sea scenario whichhas taken effective reflection and shadowing effect due to theroughness of sea surface as well as divergence and diffractionphenomena caused by earth curvature into account Butunlike the case of open-sea the influence of the obstaclesnear the inland river (eg diffraction loss caused by somelarge bridges) and mixed-path effect cannot be neglectedPaper [7] provided amethodology of field strength estimationin distance ranged from 1 km to 1000 km and frequencyranged from 30MHz to 3000MHz It is applied for land-path

HindawiInternational Journal of Antennas and PropagationVolume 2017 Article ID 5853724 15 pageshttpsdoiorg10115520175853724

2 International Journal of Antennas and Propagation

DR_X2

β

Eff_TX

Eff_RXDR_X1

D_1 D_2

RE

IN

hTX

h RX

D_LOS

tca

Figure 1 Round Earth Loss model

sea-path and mixed-path scenarios However propagationeffects including scattering divergence and earth curvaturestill have not been included either Consequently the IRmodel will take these effect factors into account and makesome quantitative analysis to improve the effectiveness andaccuracy

The rest of paper is organized as below In Section 2 ournew IRmodel is proposed based on RELmodelThree typicalmeasurement campaigns are introduced in detail in Section 3Section 4 mainly includes analysis on finite dimensionsdiffraction loss as well as study on other key propagationphenomena such as effective reflection shadowing effecton the reflected ray divergence and earth curvature effectAfterward the parameters estimation of IR model and twosimulation verifications are given in Section 5 In Section 6path loss modelsrsquo effectiveness and adaptability are evaluatedby using RMSE GRG-MAPE and PCC-MAPE algorithmsrespectively Finally conclusions are drawn in Section 7

2 Path Loss Model for Inland River

21 Round Earth Loss Model Plain Earth Loss (PEL) modelbased on Two-Ray propagation path is commonly used forcellular communication [1] However the earth cannot beregarded as a ldquoplanerdquo when the TX-RX distance exceedsseveral kilometers By accounting for scattering divergenceshadowing and earth curvature effects REL model is pro-posed for open-sea environment [6] and its geometricalmodel is given in Figure 1

The ℎTX and ℎRX are corresponding to the height ofTX and RX antennas respectively 119877119864 which is assumedas an approximate value 6731 km denotes earth radiusPropagation path loss PL119890 is given by

PL119890 = ( 1205824120587119863)2 sdot 100381610038161003816100381610038161 + 119877 sdot 119890119895(2120587120582)119863diff 100381610038161003816100381610038162 (1)

where (2120587120582) sdot 119863diff represents the phase difference betweenreflected wave and line of sight (LOS) wave 119863 is equalto 119863 LOS which denotes the length of LOS path 119877 is thespecular reflection coefficient [2 8]

119863diff = 119863119877 1198831 + 119863119877 1198832 minus 119863 (2)

= minus2

RＬＩＯＡＢSc＄ＣＰff

Effective reflectionShadowing effectDivergence

+

+

Earth curvature effect

Effect factorsPart-2

Free space lossPart-1

= 20 lg

Le_diff

20 lg + RＬＩＯＡＢ middot Sc middot ＄ＣＰff middot ej(2)Ddiff[10038161003816100381610038161003816 ]1

20 lg (

4D) = + middot 10 lg (D)

((

4)

Figure 2 The structure of REL model

where119863 is calculated by

1198632 = (119877119864 + ℎTX)2 + (119877119864 + ℎRX)2minus 2 (119877119864 + ℎTX) (119877119864 + ℎRX) cos( 119889119877119864)

(3)

The angles 120572 120573 in Figure 1 are obtained by using

120572 + 120573 = 119889119877119864 =119863 1 + 119863 2119877119864

arccos(119877119864 + ℎTX minus 119877119864 sdot cos120572119863119877 1198831 ) + 120572= arccos(119877119864 + ℎRX minus 119877119864 sdot cos120573119863119877 1198832 ) + 120573

(4)

Here the lengths of119863119877 1198831 and119863119877 1198832 are as follows119863119877 11988312 = (119877119864 + ℎTX)2 + 1198772119864 minus 2 (119877119864 + ℎTX) sdot 119877119864

sdot cos120572119863119877 11988322 = (119877119864 + ℎRX)2 + 1198772119864 minus 2 (119877119864 + ℎRX) sdot 119877119864

sdot cos120573(5)

Finally the RELmodel [6] for open-sea radio propagationcan be further expressed as follows

PL119890 = 20 lg( 1205824120587119863) + 20 lg (120578) + 119871119890 diff 120578 = 100381610038161003816100381610038161 + 119877rough sdot 119878119888 sdot Diveff sdot 119890119895(2120587120582)119863diff 10038161003816100381610038161003816

(6)

where 119877rough represents effective reflection coefficient 119878119888and Diveff are corresponding to shadowing coefficient anddivergence coefficient respectively Considering the case oflong TX-RX distance 119871119890 diff denotes the diffraction lossderived from earth curvature effect On the basis of (6) thestructure of this propagation path loss model is divided intotwo parts (shown in Figure 2)

Free space model (Part-1) estimates the path loss derivedfrom LOS ray between TX and RX Also this model can be

International Journal of Antennas and Propagation 3

W + W middot 10 lg (D)Ms = s + s middot 10 lg (D)

minus20 lg + RＬＩＯＡＢ middot Sc middot ＄ＣＰff middot ej(2)D＞Ｃff minus Le_＞Ｃff minus20 lg + RＬＩＯＡＢ middot Sc middot ＄ＣＰff middot e

j(2)D＞Ｃff minus Le_＞Ｃff

+

L + L middot 10 lg (D)20 lg (

4) minus 20 lg (D)

Ml = l + l middot 10 lg (D)+

20 lg (

4) minus 20 lg (D)

+

+

++

Land part Water partPLL PLW

Mixed-path methodology

Inland River (IR) model

] ][100381610038161003816100381610038161 [100381610038161003816100381610038161

oblm oblmObstacles diffraction loss Obstacles diffraction loss

Figure 3 The structure of Inland River model

transformed to a linear function of lg(119863) [9] For reflectedsignal various effect factors such as effective reflectionshadowing divergence and earth curvature will increase thepropagation path loss especially for the case at large TX-RXdistances for the open sea [6] Thus Part-2 is an integrationof them to make REL model applicable for the open seaenvironment

22 Inland River Model There are common features betweensea surface and river surface Therefore scattering shadow-ing and divergence still need to be taken into considerationWith an increase of TX-RX distance the diffraction losscaused by earth curvature will also occur More details will begiven in Section 4 However inland river area which containsboth land region and water region is different from the open-sea for example buildings or bridges on the river bank mayinfluence signal transmission In addition the conductivityof fresh water (120590119888 = 0001) is also different from sea water(120590119888 = 4) [8] In summary compared to the REL model thereare three improvements in the IR model

221 Parameters Optimization Low matching degree of theRELmodel and free spacemodel by comparingwith themea-surement data proves inefficiency of these classic models [1011] which indicate the necessity of model optimization Sincethe inland river region is a combination of all-land regionand all-water region the corresponding radio propagationpath loss for these single scenarios PL119871 and PL119882 are definedas the upper and lower bounds of the IR model Accordingto the Two-Ray theory environmental correction factors119872119897

and 119872119904 are employed to optimize IR model to obtain thePL119871 and PL119882 respectively Assuming that both 119872119897 and 119872119904in dB can also be depicted by linear format dependence withthe logarithmic length of LOS path (shown in Figure 3) theparameters 120572119871 120573119871 for all-land region and the parameters 120572119882120573119882 for all-water region are obtained by

120572119871 = 20 lg( 1205824120587) + 120572119897120573119871 = 120573119897 minus 2120572119882 = 20 lg( 1205824120587) + 120572119904120573119882 = 120573119904 minus 2

(7)

where 120572119897 120573119897 120572119904 and 120573119904 are empirical parametersMore detailsabout these parametersrsquo estimation will be given in Section 5

222 Diffraction Loss Caused by Obstacles near the RiverWatercourse is narrow and sinuous The buildings locatedat the river bank or some other facilities (eg bridges) maycause an additional diffraction loss Obl119898 which is ignorablein open-sea environment Thus IR model includes this partshown in Figure 3 The calculation of this parameter can befound in Section 4

223 Mixed-Path Methodology Mixed-path feature for theinland river environment makes it necessary to combine thepath losses of land region PL119871 and water region PL119882 (shown


Transmitterantenna

(a) Transmitter antenna on the iron tower in Jianghandistrict

Receiver antenna

(b) Receiver antenna at the patrol boat

Figure 4 Transmitter antenna and receiver antenna

in Figure 3) To resolve this problem a proper method whichis proposed by [7] is expressed as follows

PLIR = (1 minus 119860) sdot PL119871 + 119860 sdot PL119882 (8)

Here PLIR expresses the propagation path loss for inlandriver ldquo119860rdquo is a mixed-path interpolation factor which is givenin (9) The percentage of the water path 119865river and the landpath 119865land can be obtained from GPS position information inthe route which is also recorded in our measurement data

119860 = (1 minus (1 minus 119865river)23)119881 (9)

119881 = max [10 10 + Δ400] (10)

Δ = 119864land (119863) minus 119864river (119863) (11)

In (11) 119864land and 119864river correspond to the equivalent fieldstrength (dB) for all-land and all-water paths respectively

119864land (119863) = 1393 minus PL119871 + 20 lg (119891) 119864river (119863) = 1393 minus PL119882 + 20 lg (119891) (12)

Here 119891 denotes the carrier frequency in MHz

3 Measurement Campaigns

In our project measurement campaigns were performed forthree different scenarios described as follows

(i) Urban watercourse environment with few obstaclesbetween TX and RX

(ii) Urban watercourse environment with huge buildingsbetween TX and RX

(iii) Suburban watercourse environment

In all of these measurement campaigns ZhongxingTelecommunicationEquipmentCorporation (ZTE) provided

most of apparatuses which contain cross-polarized basestation (BS) antenna (TX) Customer Premise Equipment(CPE) Building Baseband Unit (BBU) and Remote RadioUnit (RRU) which operate in the frequency range from1447MHz to 1467MHz The transmitter end emitted 4Gsignal with 20MHz bandwidth Considering the case oflong distance transmission uplink-downlink and specialsubframe were set to configuration 1 and configuration 7[12] during the whole tests respectively In addition bothGPS position data and the information of vessel speedwere collected and saved through a Motorola mobile phone(ME525)

31 Urban Environment Measurement Campaign DescriptionFigure 4(a) shows that RRU and transmitter antenna (15 dBigain 65∘ (Az) and 8∘ (El)) were placed on the iron towerwhich was located at the roof of a tall building near theriver Therefore the height of BS antenna is equal to 7955m(including the height of iron tower 6830m the length ofTX antenna 150m and the height of shore 975m) abovethe river level after ignoring the influence of season tide Atthe receiver side (Figure 4(b)) the CPE connected to the PCvia RJ45 port can both transmit and receive the radio signalsthrough two omnidirectional antennas (35 dBi gain) Afterthat our urban measurement campaigns can be divided intodownstream part and upstream part as follows

311 Downstream Part The patrol boat sailed from a wharfbeside our base station to the Tianxingzhou Yangtze RiverBridge with the average speed of 14 knots There was noobstacle higher than TX in this direction Moreover twobridges (shown in Figure 5) which can be found between theBS and terminal point were the SecondWuhan Yangtze RiverBridge (3511 km away from the TX) and Erqi Yangtze RiverBridge (68 km away from the TX) respectively

312 Upstream Part Using the same BS and ship we com-pleted a 1501 km measurement campaign in the upstream


Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)

1km

5km

10km

Figure 5 The route of measurement campaign 1 with few obstacles between TX and RX

Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

1km

5km

10km

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)

Figure 6 The route of measurement campaign 2 with large buildings between TX and RX

direction (shown in Figure 6)The average speed was approx-imately 13 knots during the whole process It was differentfrom the downstream environment since there were somehuge buildings (average height 157m) at the distance of700m from our base station

32 Suburban Environment Measurement Campaign Descrip-tion As shown in Figure 7 this measurement campaign wasconducted in a suburban area which was positioned at Deng-jiakou Town The effective height of our BS antenna (TX)was 5300m (including the height of iron tower 5000m the


Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

1km

5km

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)

Figure 7 The route of suburban measurement campaign

length of TX antenna 150m and the height of shore 150m)above river level Due to the supervision region limitationour patrol boat with the CPE sailed only 3 km for this sce-nario

Ultimately all of the measurement information is sum-marized in Table 1

4 Effect Factors Analysis

As mentioned above the IR model contains some effectfactors which will be analyzed in the following contents

41 Finite Dimensions Diffraction Loss The diffraction losscaused by obstacles (such as huge buildings and bridges)along the river is an important effect factor Many approxi-mate methods based onmultiple knife-edges diffraction havebeen proposed to calculate its loss value (eg Bullingtonmethod [2] Epstein-Petersen method [13] Deygout method[14] ITU-R method [15] etc) To make such calculationsan assumption that the two sides of an obstacle are regardedas two thin screens of negligible thickness is necessary [15]In addition Professor Molisch [1] provided a comparisonof these different methods According to measurement cam-paigns Deygout method is an appropriate one due to inlandriver environment fulfills its requirement ldquosmall number ofscreens of different heightrdquo

In Figure 8(a) each building is simplified as two infinite-width screens which corresponds its two sides However theidealized infinite-width screen will increase the diffractionloss in most instances because the width of obstacles actuallyis finite Thus the finite-width screen diffraction given bypaper [15ndash17] for analyzing finite dimensions obstacles willbe employed in this paper The approach can be regarded as

a combination of multiple knife-edges diffraction in severaldirections (eg three-knife-edges diffraction consists of topleft side and right side as shown in Figure 8(b)) [15]

According to the description of Section 31 there is abridge which is 3511 km away from the base station Thecorresponding special finite-width obstacle diffraction can besimulated in Figure 9 Two parts diffraction loss should becalculated in this case

411 The Diffraction above the Bridge Figure 10 provides thegeometrical information based onmeasurement 1 After thatdiffraction loss 119871on is given by

119871on = minus20 lg(10038161003816100381610038161003816100381610038161003816100381612 minus 119890(119895sdot1205874)

radic2 sdot 119865 (]119891 up)100381610038161003816100381610038161003816100381610038161003816) + 11987102

119865 (]119891 up) = int]119891 up

0119890(minus119895sdot120587(11990522))119889119905

(13)

where ]119891 up is the Fresnel parameter According to [1] thescreen 02 (shown in Figure 10) is regarded as ldquomain screenrdquodue to a higher single knife-edge diffraction loss 11987102

]119891 up = 120579onradic 2120582 sdot (119889minus11199052ob1 +119882minus1119887 )

120579on = arctan(((ℎ119887 + ℎ119887th) minus ℎTX)1198891199052ob1 )+ arctan(((ℎ119887 + ℎ119887th) minus (ℎ119887 + ℎ119887th))119882119887 )

(14)

In (14) 120579on denotes the diffraction angle of screen 01 ℎ119887this the thickness of obstacle ℎ119887119882119887 1198891199052ob1 and ℎTX represent


Table 1 The measurement parameters

Parameters Data 1 Data 2 Data 3Carrier frequency 1457MHz 1457MHz 1457MHzBandwidth 20MHz 20MHz 20MHzTX power 498 dBm 498 dBm 498 dBmTX beam width (65∘ Az 8∘ El)lowast (65∘ Az 8∘ El)lowast (65∘ Az 8∘ El)lowastRX beam width Omni Omni OmniTX height 7955m 7955m 5300mRX height 34m 34m 34mTX gain 15 dBi 15 dBi 15 dBiRX gain 35 dBi 35 dBi 35 dBiMaxdistance 1106 km 1501 km 7778 kmTemperature 8ndash18∘C 8ndash18∘C 11ndash21∘CFile size 6204GB 6204GB 6204GBlowastSector antenna

TX RX

RXTX

(a) Obstacles approximation

RX

TX

(b) Finite dimensions obstacle diffraction

Figure 8 Obstacles approximation and finite dimensions obstacle diffraction

bridgersquos height bridgersquos width and distance fromTX to screen01 and the height of TX respectively

412 The Diffraction Loss under the Bridge The diffractionloss under the bridge 119871down is calculated by using

119871down = minus20 lg(10038161003816100381610038161003816100381610038161003816100381612 minus 119890(119895sdot1205874)

radic2 sdot 119865 (]119891 down)100381610038161003816100381610038161003816100381610038161003816)

+ 11989701119865 (]119891 down) = int]119891 down

0119890(minus119895sdot120587(11990522))119889119905

(15)

where ]119891 down is the Fresnel parameter in this case which canbe obtained by using (16) Unlike the diffraction part abovethe bridge the screen 01 (shown in Figure 11) results in a largerdiffraction loss 11989701 than the screen 02

]119891 down = 120579underradic 2120582 sdot (119889minus1ob2 2119903 +119882minus1119887 ) (16)

Here 119889ob2 2119903 denotes the distance between screen 02 to RXThe 120579under is given by

120579under = arctan(((ℎTX minus ℎ119887) minus (ℎTX minus ℎ119887))119882119887 )+ arctan(((ℎTX minus ℎ119887) minus (ℎTX minus ℎRX))119889ob2 2119903 )

(17)

The ℎRX which can be found in Figure 11 denotes the RXantenna height According to [15] the total bridge diffractionloss is expressed as

120585on (120599) = 10(119871on20)120585down (120599) = 10(119871down20)

Difmean (120599) = minus10 sdot lg (120585minus2on (120599) + 120585minus2down (120599)) (18)

where 120585on and 120585down are the loss factors Difmean denotesthe average diffraction loss of the bridge Since the SecondYangtzeRiver Bridge is a suspension bridge the constructionson it cannot be ignored Therefore screen 01 and screen 02


Inland river

Land

Sea level

Tower height

Distance

TX

RX

Altitude = 1755mAltitude =

hb = 2400m

Wb = 298m

hTX

=7955

m6830m

300mhbth =

dob2_2r 34mhRX =351164mdt2ob1 =

15m

975m

Figure 9 Finite dimensions bridge diffraction

Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m

6830m

01 02300mhbth =

351164mdt2ob1 =dob2_2r 34mhRX =

15m

975m

Figure 10 Diffraction over the bridge

shown in Figures 10 and 11 should also include the height ofobstacles On the basis of measurement data the diffractionloss caused by this bridge is shown in Figure 12

The results indicate that there is a large increase ofdiffraction up to 18 dB loss between 4 km and 6 km Itsmaximum value occurs at a TX-RX distance of 65 km awayfrom the BS The corresponding diffraction loss maintainswithin [0 5 dB] when the TX-RX distance is longer than9 km

In measurement campaign 2 there are some high build-ings with an average height of 157m at a TX-RX distance of700m The finite dimensions obstaclesrsquo diffraction for thisscenario can be approximated as an equivalent structure inFigure 13 Similar estimation approach is used to acquire theirdiffraction loss

As shown in Figures 14 and 15 obstacle 1 will influencethe measurement route ranging from 3 km to 10 km whileobstacle 2 will bring additional diffraction loss between 10 kmand 15 kmThus a diffraction loss dip should be found at thedistance of 10 km which is caused by a 401m space betweenobstacle 1 and obstacle 2

For measurement campaign 3 in suburban environmentthe diffraction loss can be neglected (Obl119898 = 0 dB) because

the height of base station is much higher than the otherbuildings nearby

42 Other Effect Factors IR model inherits some effectfactors from the REL model due to their similarity of envi-ronment All of them will be described in detail as follows

421 Effective Reflection from River Surface The inland riversurface is rarely smooth due to the water roughness (shownin Figure 16) As a result the scattering leads to an energyreduction compared with the idealize specular reflection [6]Assuming the height distribution of the water surface ininland river is similar with the sea wave surface which agreeswith a Gaussian distribution [18] The effective reflectioncoefficient 119877rough is given by

119877rough = 119877 sdot 119890minus2sdot((2120587120582)120590ℎ sin(1205872minus120579IN))2 EffTX = ℎTX minus 05 sdot 119877119864 sdot 1205722EffRX = ℎRX minus 05 sdot 119877119864 sdot 1205732

(19)

where 120579IN denotes the incident angle in Figure 1 EffTXand EffRX correspond to the effective height of TX and RX


01 02

Inland riverLand

Sea level

Tower height

Distance

TX

RX


hb = 2400mhTX

=7955

m

6830m

Wb = 298m

300mhbth =

351164mdt2ob1 = dob2_2r

34mhRX =

15m

975m

Figure 11 Diffraction under the bridge

5 6 7 8 9 10 114Distance (km)

minus202468

101214161820

Diff

ract

ion

loss

(dB)

Figure 12 Diffraction loss caused by bridge in measurement 1

RX

TX

Obs

tacle

2

Hei

ght

Ground1756m

401m

Hei

ght157

m

7955

m

161m

700m

877m

Obs

tacle

1

Figure 13 Finite dimensions buildings diffraction

antennas respectively 120590ℎ represents the standard deviationof surface height distribution With increasing distance ofTX and RX the grazing angle 120601 (120601 = 1205872 minus 120579IN) reducesaccordingly and approaches to zero when the radio linkis tangent to the horizon In this situation the effectivereflection coefficient 119877rough will approximate to the specularreflection coefficient 119877 According to [8] the value of 119877 isdetermined by permittivity 120576119904 and conductivity 120590119888 for variousterrain types Thus the specular reflection coefficient of the

4 5 6 7 8 9 103Distance (km)

0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)

Figure 14 Diffraction loss caused by obstacle 1 in measurement 2

0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)

10 11 12 13 14 159Distance (km)


Sea or inland river surface

Incident rayEffectively reflected ray

Scattered rays

Figure 16 Scattering and effective reflection


Specular reflection coefficient R for sea waterSpecular reflection coefficient R for fresh water

065

07

075

08

085

09

095

1

Mag

nitu

de o

f spe

cula

r re

flect

ion

coeffi

cien

t

10 15 20 25 30 35 40 455Distance (km)

Figure 17 119877 as a function of the TX-RX distance (ℎTX = 7955mℎRX = 34m)

Sea or river surface

Incident rayReflected ray

Mean plane

IN

Figure 18 Shadowing effect on the reflected ray

sea water (120576119904 = 80 120590119888 = 4) is different from fresh water (120576119904 =80 120590119888 = 0001) of inland river (shown in Figure 17)

422 Shadowing Effect on the Reflected Ray AlthoughKirch-hoff Theory assumes that any point on the surface doesnot block others [1] the shadowing may still occur duringthe measurement campaigns especially when the incidentangle 120579IN is large enough (shown in Figure 18) Accordingto [19] this phenomenon should be considered when theelevation angle 120579119905119888119886 (shown in Figure 1) is less than 05degrees Afterwards the shadowing coefficient 119878119888 based onthe assumption that the sea surface and its slope can bemodeled by using a two-dimensional Gaussian process iscalculated by

119878119888 = 1 minus 05 sdot erfc (cot 120579INradic21199080)Λ + 1 Λ = 12 (radic 2120587 sdot 1199080

cot 120579IN sdot 119890(minuscot2120579IN(2sdot11990820))

minus erfc(cot 120579INradic21199080 )) (20)

where erfc is an error function and 11990820 denotes the mean-square surface slope [20] The shadowing coefficient with

w0 = 0003

w0 = 0008

w0 = 002

383895km010203040506070809

1

Shad

owin

g co

effici

ent

05 10 15 20 25 30 35 40 450

Distance (km)


different 11990820 can be found in Figure 19 The figure illustratesthat a larger value of 1199080 will lead to a smaller shadowingcoefficient Moreover all of 119878119888 are equal to zero becausereflected rays are totally shadowed when the TX-RX distanceexceeds 383895 km

423 Divergence The earth curvature can decrease thepower density 120588119864 carried by the reflected ray as shownin Figure 20 The reduction of 120588119864 will lead to a lowerreceived signal level (RSL) [6] Consequently the divergencecoefficient Diveff also should be taken into account for long-distance inland river path loss model

Diveff

= 1

radic1 + (2 sdot 119863 1119863 2) 119877119864 (1198641198671 + 1198641198672) 1198641198671 gt 0 1198641198672 gt 00 otherwise

(21)

where 119863 1 and 119863 2 are shown in Figure 1 1198641198671 and 1198641198672represent EffTX and EffRX respectively Figure 21 indicatesthat the influence of Diveff will become bigger with theincreasing distance 119889 Ultimately divergence coefficient alsoturns to zero beyond the distance 383895 km

424 Earth Curvature Effect For the long-distance scenariothe surface of earth can be regarded as an obstacle which willblock 06 First Fresnel Zone (FFZ) 11986306(119891 ℎTX ℎRX) betweenTX and RX antennas

11986306 = 119863119891 sdot 119863ℎ119863119891 + 119863ℎ km119863119891 = 389 sdot 10minus5 sdot 119891ℎTXℎRX119863ℎ = 41 (radicℎTX + radicℎRX)

(22)

References [21 22] propose a theory of ground-wavepropagation over a smooth spherical earth (shown in Fig-ure 22) which is able to analyze the diffraction loss caused by


Figure 20 Divergence effect

5 10 15 20 25 30 35 40 450Distance (km)

0010203040506070809

1

Diff

ract

ion

loss

(dB)

383895km

Figure 21 Divergence coefficient as a function of the TX-RXdistance (ℎTX = 7955m ℎRX = 34m)

earth curvature effectThe 119889TXh denotes the distance betweenTX and the horizon while 119889RXh represents the distancebetween RX and the horizon Both of them can be obtainedby

Dis 1 = 119889TXh = radic2119896119890119877119864 sdot ℎTX

Dis 2 = 119889RXh = radic2119896119890119877119864 sdot ℎRX119889sum = Dis 1 + Dis 2

(23)

where 119896119890 is equal to 1 [19] The TX-RX distance 119889 can bedivided into Dis 1 Dis 2 and Dis 3 shown in Figure 22Thenthe corresponding diffraction loss of each part is given asfollows [6]

1198711 = 20 lg( 1198731radic56561205871205771)

1198712 = 20 lg (1198732)

RE

hTX

dTXhdRXh

h RX

Dis_1Dis_2

Dis_3

Figure 22 Ground-wave propagation over a smooth sphericalearth

20 lg (119873119899) = minus05 + 35 lg (120577119899) + 10 lg (119865119904) 120577119899 = ((2120587 sdot Dis 119899) 120582)

((2120587 sdot 119896119890 sdot 119877119864) 120582)23 119899 = 1 2 320 lg (119865119904) = minus00481205773119899 + 108751205772119899 + 40782120577119899

minus 088061198713 = 0008612057733 + 0206312057723 + 1109971205773

minus 08934(24)

Here 119865119904 and 1198713 are given by approximated polynomialfunctions from [6] After that the total path loss 119871119890 diff isrepresented as follows

If 119889 ge 119889sum 1198713 will be treated as a loss because Dis 3 ismore than 0 [6] Thus 119871119890 diff is obtained as

119871119890 diff = 1198711 + 1198712 minus 100381610038161003816100381611987131003816100381610038161003816 (25)

If 119889 lt 11986306 the diffraction loss 119871119890 diff will be equal to 0because the radio link is over the horizon

119871119890 diff = 0 (26)

In this case the earth curvature can be ignoredIf 11986306 le 119889 lt 119889sum 1198713 will be regarded as a gain owing to

the fact that 119889 is shorter than (Dis 1 + Dis 2) [6] Thus the119871119890 diff can be represented by

119871119890 diff = 1198711 + 1198712 + 100381610038161003816100381611987131003816100381610038161003816 (27)

Last but not least limited by the law of conservation ofenergy 119871119890 diff also should be set to zero when |1198713| is biggerthan |1198711| + |1198712| in this situation

Based on the above-mentioned three measurement cam-paigns 11986306 are equal to 113590 km 113590 km and80229 km respectively Therefore measurement data 1with Maxdistance 1106 km and measurement data 3 withMaxdistance 7778 km fulfill the condition 119889 lt 11986306 How-ever in measurement 2 some positions are between 11986306(113590 km) and 119889sum (383895 km) which can be found inFigure 23 It means that the diffraction loss derived fromearth curvature effect will appear from 113590 km to1501 km


minus35

minus30

minus25

minus20

minus15

minus10

minus5

0

Diff

ract

ion

loss

(dB)

5 10 15 20 25 30 35 40 450Distance (km)

Measurement-2d = 1501 km

D06 = 113590km

dsum = 383895km

Figure 23 Diffraction loss caused by earth curvature in the case ofmeasurement data 2

ITU-R mixed-path 50ITU-R mixed-path 10ITU-R mixed-path 1ITU-R sea-path 50ITU-R sea-path 10ITU-R sea-path 1ITU-R land-path 50

60

80

100

120

140

160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)

ITU-R land-path 10ITU-R land-path 1Okumura-Hata suburbanOkumura-Hata urbanRELFree spaceMeasurement data

Figure 24 The simulation of classic radio propagation path lossmodels based on measurement data 1

5 Parameters Estimation andSimulation Verification

As mentioned in Section 2 IR model has four empiricalparameters 120572119897 120573119897 120572119904 and 120573119904 Figure 24 shows a series ofclassic radio propagation path loss models which are widelyused in wireless communication such as free space modelREL model Okumura-Hata model and ITU-R modelsThe simulation outcomes indicate that measurement resultswhich are between ITU-R land-path model and sea-pathmodel show obvious mixed-path characteristics HoweverITU-R mixed-path models do not have a good fitting degreewhich is displayed in Figure 24 Comparedwith othermodelsOkumura-Hata urban model and suburban model matchbetter with the raw data In addition both of them are able tobe expressed in logarithmic scale [3 4] We assume that theenvironmental correction factor (shown in Section 22 andFigure 3) for all-water regions119872119904 can be depicted by (28)

119872119904 = 120572sub minus 20 lg( 1205824120587) + [120573sub + 2] 10 lg (119863) (28)

110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)

Diffraction losscaused by bridge

Reflection effect

IR model w20 = 0002

IR model w20 = 0004

IR model w20 = 0008

Measurement dataIR model w2

0 = 002

Figure 25 IR model with different 11990820

120572119904 = 120572sub minus 20 lg( 1205824120587) 120573119904 = 120573sub + 2

(29)

Here 120572sub and 120573sub are derived from Okumura-Hatasuburban model [3 4] For all-land regions the Okumura-Hata urban model and suburban model are employed tocalculate the corresponding correction factors119872119897 (urban orsuburban)

In the case of urban environment

119872119897 = 120572urb minus 20 lg( 1205824120587) + [120573urb + 2] 10 lg (119863) 120572119897 = 120572urb minus 20 lg( 1205824120587)

120573119897 = 120573urb + 2

(30)

In the case of suburban environment

119872119897 = 119872119904 (31)

Here 120572urb and 120573urb can be acquired from the Okumura-Hata model for urban scenario [3 4] After that the IRmodel with different standard deviation of surface heightdistribution 120590ℎ andmean-square surface slope11990820 (the valuesof other variables are the same) are described in Figures 25and 26 respectively It can be found that the smaller 11990820 and120590ℎ will lead to the higher loss peaks derived from reflectioneffect On the basis of the comparison outcomes IR modelchooses 0002 and 02 as the values of 11990820 and 120590ℎ which alsoconfirms that the surface of river is smoother than open-sea(11990820 = 025 120590ℎ = 0008 in [6]) Moreover a good matchingdegree of diffraction loss proves that the finite dimensionsbridge diffraction in IRmodel are applicable for this scenario

To verify the universal applicability of IR model mea-surement data 2 and measurement data 3 are employed asreference sequences Figure 27 shows that IR model agrees


110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)


Reflection effect

2 3 4 5 6 7 8 9 10 111Distance (km)

Measurement dataIR modelIR model h = 075

IR model h = 05

IR model h = 025

h = 02

Figure 26 IR model with different 120590ℎ

4 6 8 10 12 142Distance (km)

60

80

100

120

140

160

180

Prop

agat

ion

path

loss

(dB)

Region 1 Region 2 Region 3

Reflection effect

ITU-R mixed-path 50ITU-R mixed-path 10ITU-R mixed-path 1ITU-R sea-path 50ITU-R seat-path 10ITU-R sea-path 1ITU-R land-path 50ITU-R land-path 10

ITU-R land-path 1Okumura-Hata suburbanOkumura-Hata urbanRELFree space

Measurement dataIR model

Figure 27 Simulation verification based on measurement data 2Region 1 denotes the diffraction loss caused by obstacle 1 The dip inRegion 2 is derived from space between the two buildings Region 3represents diffraction loss caused by obstacle 2

well with the reflection phenomenon and the diffraction losscaused by huge buildings near the river which have greateffect on signal transmission In addition as mentioned inSection 41 the dip occurring around the 10 km is due toa 401m space between the two buildings In Figure 28both the Okumura-Hata (suburban) model and the IRmodelmatch well with the measurement data In conclusion thesecomparisons for the three scenarios prove that the inlandriver scenario can be approximated as a mixed-path Two-Ray geometric model Secondly the assumptions of param-eters optimization in Section 2 and parameters estimationin Section 5 are reasonable Thirdly the finite dimensionsmethod is verified to be an effective way to calculate thediffraction loss caused by bridges huge buildings and someother facilities near the river As shown in Figures 25 26

60708090

100110120130140150160

Prop

agat

ion

path

loss

(dB)

55 6 65 7 755Distance (km)

ITU-R mixed-path 50ITU-R mixed-path 10ITU-R mixed-path 1ITU-R sea-path 50ITU-R sea-path 10ITU-R sea-path 1ITU-R land-path 50ITU-R land-path 10



Figure 28 Simulation verification based on measurement data 3

and 27 reflection phenomenon is the dominant influencefactor from 1 km to 3 km which will bring a loss withinthe scope of [0 15] dB On the basis of quantitative analysisin Section 3 diffraction phenomenon is more obvious thanother phenomena in 4 kmsim6 km of measurement 1 and3 kmsim15 km of measurement 2

6 Performance Evaluation

In order to quantify the practicability of these models threekinds of model selection algorithms are employed RMSEcriteria is a commonly used methodology to evaluate thedifference between the theoretical model and measurementdata [11 23 24] due to low complexity By defining the modelselection as an ldquotrack association problemrdquo in automation andtraffic field [10] provides two newmethods (GRG-MAPE andPCC-MAPE) based on Uncertainty-Mathematical Theorywhich have been proven more accurate than RMSE

On the basis of the three methodologies mentionedabove Tables 2 3 and 4 list the performance evaluationresults of path loss models respectively

In Table 2 RMSE values denote the error levels of theestimation sequences Thus both free space model and RELmodel are not appropriate for all of these three scenariosdue to high RMSE values As shown in Data 1 and Data 2Okumura-Hata (urban) model shows smaller RMSE values(78181 and 80515) than the suburban model and the ITU-R models apart from the IR model (46159 and 63594)For suburban watercourse scenario (Data 3) IR model andOkumura-Hata (suburban)model are the optimal model andthe suboptimalmodel respectivelyThe values in Tables 3 and4 describe the matching degrees of corresponding path lossmodels Afterwards all of comparison results indicate thatIR model performs most effectively for these environmentswhereas the Okumura-Hata (suburban) model also matcheswell with our measurement data in suburban scenario


Table 2 The RMSE model selection results

Models MeasurementsData 1lowast Data 2lowast Data 3lowast

ITU-R land-path 1 137119 93384 75479ITU-R land-path 10 143341 93848 91265ITU-R land-path 50 146268 94386 97505ITU-R sea-path 1 210225 303053 262342ITU-R sea-path 10 196629 287838 250719ITU-R sea-path 50 187016 277185 24146ITU-R mixed-path 1 144477 151869 150571ITU-R mixed-path 10 132697 141991 136386ITU-R mixed-path 50 125827 137837 129074Free space model 304055 425521 254202Okumura-Hata (suburban) 92825 188068 18525Okumura-Hata (urban) 78181 80515 129709REL model 333084 442648 282632IR model 46159 63594 18310lowastThe smaller RMSE value means better matching degree

Table 3 The GRG-MAPE model selection results


ITU-R land-path 1 08833 09315 09264ITU-R land-path 10 08795 09318 09138ITU-R land-path 50 08774 09317 09094ITU-R sea-path 1 08648 08090 08064ITU-R sea-path 10 08725 08175 08137ITU-R sea-path 50 08779 08235 08193ITU-R mixed-path 1 09136 08976 08778ITU-R mixed-path 10 09201 09030 08892ITU-R mixed-path 50 09240 09054 08944Free space model 07865 07360 08123Okumura-Hata (suburban) 09296 08774 09570Okumura-Hata (urban) 09303 09378 08934REL model 07659 07206 07853IR model 09574 09522 09582lowastThe larger GRG-MAPE value means better matching degree

For the computational complexity of these path lossmod-els free space model [1] Okumura-Hata suburban modeland its urban model [3 4] are much simpler than others dueto lack of sufficient effect factors analysis ITU-R models [7]own many corrections to the prediction results includingterrain clearance angle correction antenna height mixed-path and tropospheric scattering However these commonlyused classic empirical models do not analyze deterministicinformation which is calculated in IR model such as diffrac-tion loss caused by bridges or buildings Moreover IR modelhas three improvements compared with original REL model[6] which can be found in Section 2 Consequently IRmodelis more complex than ITU-R models and REL model

Table 4 The PCC-MAPE model selection results


ITU-R land-path 1 08891 09321 09419ITU-R land-path 10 08848 09324 09317ITU-R land-path 50 08832 09329 09276ITU-R sea-path 1 08773 08165 08195ITU-R sea-path 10 08838 08246 08271ITU-R sea-path 50 08884 08305 08329ITU-R mixed-path 1 09263 09046 08932ITU-R mixed-path 10 09318 09102 09024ITU-R mixed-path 50 09352 09127 09070Free space model 07967 07447 08242Okumura-Hata (suburban) 09390 08851 09797Okumura-Hata (urban) 09398 09458 09058REL model 07581 07173 08058IR model 09688 09612 09822lowastThe larger PCC-MAPE value means better matching degree

7 Conclusion

In this paper a new propagation path loss model based onthe REL model is proposed for inland river environmentCompared with the REL model this model shows threeimprovements derived from differences between open-seaand inland river In the aspect of parameters optimizationthe formula in dB scale transformed from the free spacemodel is replaced by the Okumura-Hata model Secondlyan additional diffraction loss caused by the obstacles nearthe river has been included Then the mixed-path fea-ture is taken into consideration by using the methodologyfrom Recommendation ITU-R P1546-5 To acquire referencegroups and perform parameters estimation three measure-ment campaigns have been conducted for different scenariosAccording to effect factors analysis and simulation resultsthe high matching degree of IR model indicates that theradio channel of inland river scenario can be simplifiedby using Two-Ray geometric model and the method oflinear parameters optimization as well as the calculation offinite dimensions diffraction loss are effective Besides thecomparison between the measurement results and classicalpath loss models suggests that propagation path withoutmixture of land and water path is not appropriate for inlandriver owing to its special mixed-path characteristics In orderto better evaluate the modelrsquos effectiveness RMSE GRG-MAPE and PCC-MAPE are used for model selection Thecorresponding outcomes also prove that our IR model whichconsiders mixed path and some deterministic informationmatches best with the measurement results

Competing Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper


Acknowledgments

The authors would like to express their sincere thanks tothe ZTE for their generous equipment support Super RadioAS for technical support on data analysis and ChangjiangMaritime Safety Administration for their coordination workduring the measurement process The work of Kun Yang wassupported by the MAMIME project under Grant Agreementno 256309 by Norwegian Research Council

References

[1] A F Molisch Wireless Communications John Wiley amp SonsNew York NY USA 2007

[2] J D Parsons The Mobile Radio Propagation Channel JohnWiley amp Sons New York NY USA 2000

[3] Y Okumura E Ohmori and T Kawano ldquoField strength and itsvariability in VHF and UHF land-mobile radio servicerdquo Reviewof the Electrical Communication Laboratory vol 16 no 9 pp825ndash873 1968

[4] MHata ldquoEmpirical formula for propagation loss in landmobileradio servicesrdquo IEEE Transactions on Vehicular Technology vol29 no 3 pp 317ndash325 1980

[5] L Klozar and J Prokopec ldquoPropagation path loss models formobile communicationrdquo in Proceedings of the 21st InternationalConference Radioelektronika pp 287ndash290 IEEE Brno CzechRepublic April 2011

[6] K Yang A F Molisch T Ekman and T Roslashste ldquoA deterministicround earth loss model for open-sea radio propagationrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC Spring rsquo13) pp 1ndash5 IEEE Dresden Germany June 2013

[7] Recommendation ITU-R P1546-5 ldquoMethod for point-toareapredictions for terrestrial services in the frequency range30MHz to 3000MHzrdquo September 2013

[8] N H Lu ldquoLinearized unified two-ray formulation for propaga-tion over a plane earthrdquo in Proceedings of the Sensors for IndustryConference pp 95ndash100 IEEE Houston Tex USA February2005

[9] M R Akdeniz Y Liu M K Samimi et al ldquoMillimeter wavechannel modeling and cellular capacity evaluationrdquo IEEE Jour-nal on SelectedAreas in Communications vol 32 no 6 pp 1164ndash1179 2014

[10] J Yu C Li K Yang and W Chen ldquoGRG-MAPE and PCC-MAPE based on uncertainty-mathematical theory for path-loss model selectionrdquo in Proceedings of the 83rd IEEE VehicularTechnology Conference (VTC Spring rsquo16) May 2016

[11] C Li J Yu K Yang and W Chen ldquoGrey System Theory app-lication in model selection of channel modeling at 14 GHzrdquoin Proceedings of the International Conference on WirelessCommunications Signal Processing and Networking (WiSPNETrsquo16) pp 1745ndash1749 IEEE Chennai India March 2016

[12] 3GPP ldquoPhysical channels and modulationrdquo 3GPP TS 36211Technical Specification Group Radio Access Network

[13] J Epstein and D W Peterson ldquoAn experimental study of wavepropagation at 850 MCrdquo Proceedings of the IRE vol 41 no 5pp 595ndash611 1953

[14] J Deygout ldquoMultiple knife-edge diffraction of microwavesrdquoIEEE Transactions on Antennas and Propagation vol 14 no 4pp 480ndash489 1966

[15] Recommendation ITU-RP526-13 ldquoPropagation by diffractionrdquoNovember 2013

[16] E Torabi A Ghorbani and A Tajvidy ldquoA modified diffractioncoefficient for imperfect conducting wedges and buildingswith finite dimensionsrdquo IEEE Transactions on Antennas andPropagation vol 57 no 4 pp 1197ndash1207 2009

[17] M B Raza andT Fickenscher ldquoDiffraction loss and phasemod-ulation of terrestrial radio-link by wind turbinerdquo in Proceedingsof the International Workshop on Antenna Technology SmallAntennas Novel EM Structures and Materials and Applications(iWAT rsquo14) pp 382ndash384 Sydney Australia March 2014

[18] K Yang T Ekman T Roste et al ldquoA quasi-deterministic pathloss propagationmodelfortheopenseaenvironmentrdquo in Proceed-ings of the IEEE 14th International Symposium on WirelessPersonal Multimedia Communications (WPMC rsquo11) pp 1ndash5October 2011

[19] K Yang Channel measurements and channel modeling for theopen sea [PhD thesis] Norwegian University of Science andTechnology (NTNU) Trondheim Norway 2013

[20] B Smith ldquoGeometrical shadowing of a random rough surfacerdquoIEEE Transactions on Antennas and Propagation vol 15 no 5pp 668ndash671 1967

[21] K Bullington ldquoRadio propagation at frequencies above 30megacyclesrdquo Proceedings of the IRE vol 35 no 10 pp 1122ndash11361947

[22] K Yang T Roste F Bekkadal K Husby and O TrandemldquoLong-distance propagation measurements of mobile radiochannel over sea at 2 GHzrdquo in Proceedings of the IEEE VehicularTechnology Conference (VTC Fall rsquo11) September 2011

[23] M Farhoud A El-Keyi and A Sultan ldquoEmpirical correctionof the Okumura-Hata model for the 900 MHz band in Egyptrdquoin Proceedings of the 3rd International Conference on Commu-nications and Information Technology (ICCIT rsquo13) pp 386ndash390IEEE June 2013

[24] I RodriguezHCNguyen T B Sorensen et al ldquoA geometrical-based vertical gain correction for signal strength prediction ofdowntilted base station antennas in urban areasrdquo in Proceedingsof the 76th IEEE Vehicular Technology Conference (VTC rsquo12) pp1ndash5 IEEE Quebec Canada September 2012

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpswwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



DR_X2

β

Eff_TX

Eff_RXDR_X1

D_1 D_2

RE

IN

hTX

h RX

D_LOS

tca

Figure 1 Round Earth Loss model

sea-path and mixed-path scenarios However propagationeffects including scattering divergence and earth curvaturestill have not been included either Consequently the IRmodel will take these effect factors into account and makesome quantitative analysis to improve the effectiveness andaccuracy

The rest of paper is organized as below In Section 2 ournew IRmodel is proposed based on RELmodelThree typicalmeasurement campaigns are introduced in detail in Section 3Section 4 mainly includes analysis on finite dimensionsdiffraction loss as well as study on other key propagationphenomena such as effective reflection shadowing effecton the reflected ray divergence and earth curvature effectAfterward the parameters estimation of IR model and twosimulation verifications are given in Section 5 In Section 6path loss modelsrsquo effectiveness and adaptability are evaluatedby using RMSE GRG-MAPE and PCC-MAPE algorithmsrespectively Finally conclusions are drawn in Section 7

2 Path Loss Model for Inland River

21 Round Earth Loss Model Plain Earth Loss (PEL) modelbased on Two-Ray propagation path is commonly used forcellular communication [1] However the earth cannot beregarded as a ldquoplanerdquo when the TX-RX distance exceedsseveral kilometers By accounting for scattering divergenceshadowing and earth curvature effects REL model is pro-posed for open-sea environment [6] and its geometricalmodel is given in Figure 1

The ℎTX and ℎRX are corresponding to the height ofTX and RX antennas respectively 119877119864 which is assumedas an approximate value 6731 km denotes earth radiusPropagation path loss PL119890 is given by

PL119890 = ( 1205824120587119863)2 sdot 100381610038161003816100381610038161 + 119877 sdot 119890119895(2120587120582)119863diff 100381610038161003816100381610038162 (1)

where (2120587120582) sdot 119863diff represents the phase difference betweenreflected wave and line of sight (LOS) wave 119863 is equalto 119863 LOS which denotes the length of LOS path 119877 is thespecular reflection coefficient [2 8]

119863diff = 119863119877 1198831 + 119863119877 1198832 minus 119863 (2)

= minus2

RＬＩＯＡＢSc＄ＣＰff

Effective reflectionShadowing effectDivergence

+

+

Earth curvature effect

Effect factorsPart-2

Free space lossPart-1

= 20 lg

Le_diff

20 lg + RＬＩＯＡＢ middot Sc middot ＄ＣＰff middot ej(2)Ddiff[10038161003816100381610038161003816 ]1

20 lg (

4D) = + middot 10 lg (D)

((

4)

Figure 2 The structure of REL model

where119863 is calculated by

1198632 = (119877119864 + ℎTX)2 + (119877119864 + ℎRX)2minus 2 (119877119864 + ℎTX) (119877119864 + ℎRX) cos( 119889119877119864)

(3)

The angles 120572 120573 in Figure 1 are obtained by using

120572 + 120573 = 119889119877119864 =119863 1 + 119863 2119877119864

arccos(119877119864 + ℎTX minus 119877119864 sdot cos120572119863119877 1198831 ) + 120572= arccos(119877119864 + ℎRX minus 119877119864 sdot cos120573119863119877 1198832 ) + 120573

(4)

Here the lengths of119863119877 1198831 and119863119877 1198832 are as follows119863119877 11988312 = (119877119864 + ℎTX)2 + 1198772119864 minus 2 (119877119864 + ℎTX) sdot 119877119864

sdot cos120572119863119877 11988322 = (119877119864 + ℎRX)2 + 1198772119864 minus 2 (119877119864 + ℎRX) sdot 119877119864

sdot cos120573(5)

Finally the RELmodel [6] for open-sea radio propagationcan be further expressed as follows

PL119890 = 20 lg( 1205824120587119863) + 20 lg (120578) + 119871119890 diff 120578 = 100381610038161003816100381610038161 + 119877rough sdot 119878119888 sdot Diveff sdot 119890119895(2120587120582)119863diff 10038161003816100381610038161003816

(6)

where 119877rough represents effective reflection coefficient 119878119888and Diveff are corresponding to shadowing coefficient anddivergence coefficient respectively Considering the case oflong TX-RX distance 119871119890 diff denotes the diffraction lossderived from earth curvature effect On the basis of (6) thestructure of this propagation path loss model is divided intotwo parts (shown in Figure 2)

Free space model (Part-1) estimates the path loss derivedfrom LOS ray between TX and RX Also this model can be





+


4) minus 20 lg (D)


20 lg (

4) minus 20 lg (D)

+

+

++




] ][100381610038161003816100381610038161 [100381610038161003816100381610038161







120572119871 = 20 lg( 1205824120587) + 120572119897120573119871 = 120573119897 minus 2120572119882 = 20 lg( 1205824120587) + 120572119904120573119882 = 120573119904 minus 2

(7)





Transmitterantenna


Receiver antenna







119881 = max [10 10 + Δ400] (10)
















Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)

1km

5km

10km


Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

1km

5km

10km

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)





Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

1km

5km

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)












radic2 sdot 119865 (]119891 up)100381610038161003816100381610038161003816100381610038161003816) + 11987102

119865 (]119891 up) = int]119891 up

0119890(minus119895sdot120587(11990522))119889119905

(13)




(14)





TX RX

RXTX


RX

TX






radic2 sdot 119865 (]119891 down)100381610038161003816100381610038161003816100381610038161003816)

+ 11989701119865 (]119891 down) = int]119891 down

0119890(minus119895sdot120587(11990522))119889119905

(15)





(17)


120585on (120599) = 10(119871on20)120585down (120599) = 10(119871down20)




Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m6830m

300mhbth =


15m

975m


Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m

6830m

01 02300mhbth =


15m

975m











(19)



01 02

Inland riverLand

Sea level

Tower height

Distance

TX

RX


hb = 2400mhTX

=7955

m

6830m

Wb = 298m

300mhbth =

351164mdt2ob1 = dob2_2r

34mhRX =

15m

975m


5 6 7 8 9 10 114Distance (km)

minus202468

101214161820

Diff

ract

ion

loss

(dB)


RX

TX

Obs

tacle

2

Hei

ght

Ground1756m

401m

Hei

ght157

m

7955

m

161m

700m

877m

Obs

tacle

1



4 5 6 7 8 9 103Distance (km)

0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)


0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)

10 11 12 13 14 159Distance (km)




Scattered rays




065

07

075

08

085

09

095

1

Mag

nitu

de o

f spe

cula

r re

flect

ion

coeffi

cien

t

10 15 20 25 30 35 40 455Distance (km)




Mean plane

IN








w0 = 0003

w0 = 0008

w0 = 002

383895km010203040506070809

1

Shad

owin

g co

effici

ent

05 10 15 20 25 30 35 40 450

Distance (km)




Diveff

= 1


(21)




(22)




5 10 15 20 25 30 35 40 450Distance (km)

0010203040506070809

1

Diff

ract

ion

loss

(dB)

383895km





(23)


1198711 = 20 lg( 1198731radic56561205871205771)

1198712 = 20 lg (1198732)

RE

hTX

dTXhdRXh

h RX

Dis_1Dis_2

Dis_3




minus 088061198713 = 0008612057733 + 0206312057723 + 1109971205773

minus 08934(24)



119871119890 diff = 1198711 + 1198712 minus 100381610038161003816100381611987131003816100381610038161003816 (25)


119871119890 diff = 0 (26)



119871119890 diff = 1198711 + 1198712 + 100381610038161003816100381611987131003816100381610038161003816 (27)




minus35

minus30

minus25

minus20

minus15

minus10

minus5

0

Diff

ract

ion

loss

(dB)

5 10 15 20 25 30 35 40 450Distance (km)


D06 = 113590km

dsum = 383895km



60

80

100

120

140

160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)






110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)


Reflection effect

IR model w20 = 0002

IR model w20 = 0004

IR model w20 = 0008


0 = 002


120572119904 = 120572sub minus 20 lg( 1205824120587) 120573119904 = 120573sub + 2

(29)




120573119897 = 120573urb + 2

(30)


119872119897 = 119872119904 (31)




110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)


Reflection effect

2 3 4 5 6 7 8 9 10 111Distance (km)


IR model h = 05

IR model h = 025

h = 02


4 6 8 10 12 142Distance (km)

60

80

100

120

140

160

180

Prop

agat

ion

path

loss

(dB)


Reflection effect






60708090

100110120130140150160

Prop

agat

ion

path

loss

(dB)

55 6 65 7 755Distance (km)





















7 Conclusion


Competing Interests



Acknowledgments


References



























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation













+


4) minus 20 lg (D)


20 lg (

4) minus 20 lg (D)

+

+

++




] ][100381610038161003816100381610038161 [100381610038161003816100381610038161







120572119871 = 20 lg( 1205824120587) + 120572119897120573119871 = 120573119897 minus 2120572119882 = 20 lg( 1205824120587) + 120572119904120573119882 = 120573119904 minus 2

(7)





Transmitterantenna


Receiver antenna







119881 = max [10 10 + Δ400] (10)
















Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)

1km

5km

10km


Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

1km

5km

10km

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)





Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

1km

5km

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)












radic2 sdot 119865 (]119891 up)100381610038161003816100381610038161003816100381610038161003816) + 11987102

119865 (]119891 up) = int]119891 up

0119890(minus119895sdot120587(11990522))119889119905

(13)




(14)





TX RX

RXTX


RX

TX






radic2 sdot 119865 (]119891 down)100381610038161003816100381610038161003816100381610038161003816)

+ 11989701119865 (]119891 down) = int]119891 down

0119890(minus119895sdot120587(11990522))119889119905

(15)





(17)


120585on (120599) = 10(119871on20)120585down (120599) = 10(119871down20)




Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m6830m

300mhbth =


15m

975m


Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m

6830m

01 02300mhbth =


15m

975m











(19)



01 02

Inland riverLand

Sea level

Tower height

Distance

TX

RX


hb = 2400mhTX

=7955

m

6830m

Wb = 298m

300mhbth =

351164mdt2ob1 = dob2_2r

34mhRX =

15m

975m


5 6 7 8 9 10 114Distance (km)

minus202468

101214161820

Diff

ract

ion

loss

(dB)


RX

TX

Obs

tacle

2

Hei

ght

Ground1756m

401m

Hei

ght157

m

7955

m

161m

700m

877m

Obs

tacle

1



4 5 6 7 8 9 103Distance (km)

0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)


0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)

10 11 12 13 14 159Distance (km)




Scattered rays




065

07

075

08

085

09

095

1

Mag

nitu

de o

f spe

cula

r re

flect

ion

coeffi

cien

t

10 15 20 25 30 35 40 455Distance (km)




Mean plane

IN








w0 = 0003

w0 = 0008

w0 = 002

383895km010203040506070809

1

Shad

owin

g co

effici

ent

05 10 15 20 25 30 35 40 450

Distance (km)




Diveff

= 1


(21)




(22)




5 10 15 20 25 30 35 40 450Distance (km)

0010203040506070809

1

Diff

ract

ion

loss

(dB)

383895km





(23)


1198711 = 20 lg( 1198731radic56561205871205771)

1198712 = 20 lg (1198732)

RE

hTX

dTXhdRXh

h RX

Dis_1Dis_2

Dis_3




minus 088061198713 = 0008612057733 + 0206312057723 + 1109971205773

minus 08934(24)



119871119890 diff = 1198711 + 1198712 minus 100381610038161003816100381611987131003816100381610038161003816 (25)


119871119890 diff = 0 (26)



119871119890 diff = 1198711 + 1198712 + 100381610038161003816100381611987131003816100381610038161003816 (27)




minus35

minus30

minus25

minus20

minus15

minus10

minus5

0

Diff

ract

ion

loss

(dB)

5 10 15 20 25 30 35 40 450Distance (km)


D06 = 113590km

dsum = 383895km



60

80

100

120

140

160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)






110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)


Reflection effect

IR model w20 = 0002

IR model w20 = 0004

IR model w20 = 0008


0 = 002


120572119904 = 120572sub minus 20 lg( 1205824120587) 120573119904 = 120573sub + 2

(29)




120573119897 = 120573urb + 2

(30)


119872119897 = 119872119904 (31)




110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)


Reflection effect

2 3 4 5 6 7 8 9 10 111Distance (km)


IR model h = 05

IR model h = 025

h = 02


4 6 8 10 12 142Distance (km)

60

80

100

120

140

160

180

Prop

agat

ion

path

loss

(dB)


Reflection effect






60708090

100110120130140150160

Prop

agat

ion

path

loss

(dB)

55 6 65 7 755Distance (km)





















7 Conclusion


Competing Interests



Acknowledgments


References



























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Transmitterantenna


Receiver antenna







119881 = max [10 10 + Δ400] (10)
















Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)

1km

5km

10km


Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

1km

5km

10km

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)





Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

1km

5km

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)












radic2 sdot 119865 (]119891 up)100381610038161003816100381610038161003816100381610038161003816) + 11987102

119865 (]119891 up) = int]119891 up

0119890(minus119895sdot120587(11990522))119889119905

(13)




(14)





TX RX

RXTX


RX

TX






radic2 sdot 119865 (]119891 down)100381610038161003816100381610038161003816100381610038161003816)

+ 11989701119865 (]119891 down) = int]119891 down

0119890(minus119895sdot120587(11990522))119889119905

(15)





(17)


120585on (120599) = 10(119871on20)120585down (120599) = 10(119871down20)




Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m6830m

300mhbth =


15m

975m


Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m

6830m

01 02300mhbth =


15m

975m











(19)



01 02

Inland riverLand

Sea level

Tower height

Distance

TX

RX


hb = 2400mhTX

=7955

m

6830m

Wb = 298m

300mhbth =

351164mdt2ob1 = dob2_2r

34mhRX =

15m

975m


5 6 7 8 9 10 114Distance (km)

minus202468

101214161820

Diff

ract

ion

loss

(dB)


RX

TX

Obs

tacle

2

Hei

ght

Ground1756m

401m

Hei

ght157

m

7955

m

161m

700m

877m

Obs

tacle

1



4 5 6 7 8 9 103Distance (km)

0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)


0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)

10 11 12 13 14 159Distance (km)




Scattered rays




065

07

075

08

085

09

095

1

Mag

nitu

de o

f spe

cula

r re

flect

ion

coeffi

cien

t

10 15 20 25 30 35 40 455Distance (km)




Mean plane

IN








w0 = 0003

w0 = 0008

w0 = 002

383895km010203040506070809

1

Shad

owin

g co

effici

ent

05 10 15 20 25 30 35 40 450

Distance (km)




Diveff

= 1


(21)




(22)




5 10 15 20 25 30 35 40 450Distance (km)

0010203040506070809

1

Diff

ract

ion

loss

(dB)

383895km





(23)


1198711 = 20 lg( 1198731radic56561205871205771)

1198712 = 20 lg (1198732)

RE

hTX

dTXhdRXh

h RX

Dis_1Dis_2

Dis_3




minus 088061198713 = 0008612057733 + 0206312057723 + 1109971205773

minus 08934(24)



119871119890 diff = 1198711 + 1198712 minus 100381610038161003816100381611987131003816100381610038161003816 (25)


119871119890 diff = 0 (26)



119871119890 diff = 1198711 + 1198712 + 100381610038161003816100381611987131003816100381610038161003816 (27)




minus35

minus30

minus25

minus20

minus15

minus10

minus5

0

Diff

ract

ion

loss

(dB)

5 10 15 20 25 30 35 40 450Distance (km)


D06 = 113590km

dsum = 383895km



60

80

100

120

140

160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)






110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)


Reflection effect

IR model w20 = 0002

IR model w20 = 0004

IR model w20 = 0008


0 = 002


120572119904 = 120572sub minus 20 lg( 1205824120587) 120573119904 = 120573sub + 2

(29)




120573119897 = 120573urb + 2

(30)


119872119897 = 119872119904 (31)




110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)


Reflection effect

2 3 4 5 6 7 8 9 10 111Distance (km)


IR model h = 05

IR model h = 025

h = 02


4 6 8 10 12 142Distance (km)

60

80

100

120

140

160

180

Prop

agat

ion

path

loss

(dB)


Reflection effect






60708090

100110120130140150160

Prop

agat

ion

path

loss

(dB)

55 6 65 7 755Distance (km)





















7 Conclusion


Competing Interests



Acknowledgments


References



























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)

1km

5km

10km


Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

1km

5km

10km

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)





Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

1km

5km

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)












radic2 sdot 119865 (]119891 up)100381610038161003816100381610038161003816100381610038161003816) + 11987102

119865 (]119891 up) = int]119891 up

0119890(minus119895sdot120587(11990522))119889119905

(13)




(14)





TX RX

RXTX


RX

TX






radic2 sdot 119865 (]119891 down)100381610038161003816100381610038161003816100381610038161003816)

+ 11989701119865 (]119891 down) = int]119891 down

0119890(minus119895sdot120587(11990522))119889119905

(15)





(17)


120585on (120599) = 10(119871on20)120585down (120599) = 10(119871down20)




Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m6830m

300mhbth =


15m

975m


Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m

6830m

01 02300mhbth =


15m

975m











(19)



01 02

Inland riverLand

Sea level

Tower height

Distance

TX

RX


hb = 2400mhTX

=7955

m

6830m

Wb = 298m

300mhbth =

351164mdt2ob1 = dob2_2r

34mhRX =

15m

975m


5 6 7 8 9 10 114Distance (km)

minus202468

101214161820

Diff

ract

ion

loss

(dB)


RX

TX

Obs

tacle

2

Hei

ght

Ground1756m

401m

Hei

ght157

m

7955

m

161m

700m

877m

Obs

tacle

1



4 5 6 7 8 9 103Distance (km)

0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)


0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)

10 11 12 13 14 159Distance (km)




Scattered rays




065

07

075

08

085

09

095

1

Mag

nitu

de o

f spe

cula

r re

flect

ion

coeffi

cien

t

10 15 20 25 30 35 40 455Distance (km)




Mean plane

IN








w0 = 0003

w0 = 0008

w0 = 002

383895km010203040506070809

1

Shad

owin

g co

effici

ent

05 10 15 20 25 30 35 40 450

Distance (km)




Diveff

= 1


(21)




(22)




5 10 15 20 25 30 35 40 450Distance (km)

0010203040506070809

1

Diff

ract

ion

loss

(dB)

383895km





(23)


1198711 = 20 lg( 1198731radic56561205871205771)

1198712 = 20 lg (1198732)

RE

hTX

dTXhdRXh

h RX

Dis_1Dis_2

Dis_3




minus 088061198713 = 0008612057733 + 0206312057723 + 1109971205773

minus 08934(24)



119871119890 diff = 1198711 + 1198712 minus 100381610038161003816100381611987131003816100381610038161003816 (25)


119871119890 diff = 0 (26)



119871119890 diff = 1198711 + 1198712 + 100381610038161003816100381611987131003816100381610038161003816 (27)




minus35

minus30

minus25

minus20

minus15

minus10

minus5

0

Diff

ract

ion

loss

(dB)

5 10 15 20 25 30 35 40 450Distance (km)


D06 = 113590km

dsum = 383895km



60

80

100

120

140

160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)






110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)


Reflection effect

IR model w20 = 0002

IR model w20 = 0004

IR model w20 = 0008


0 = 002


120572119904 = 120572sub minus 20 lg( 1205824120587) 120573119904 = 120573sub + 2

(29)




120573119897 = 120573urb + 2

(30)


119872119897 = 119872119904 (31)




110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)


Reflection effect

2 3 4 5 6 7 8 9 10 111Distance (km)


IR model h = 05

IR model h = 025

h = 02


4 6 8 10 12 142Distance (km)

60

80

100

120

140

160

180

Prop

agat

ion

path

loss

(dB)


Reflection effect






60708090

100110120130140150160

Prop

agat

ion

path

loss

(dB)

55 6 65 7 755Distance (km)





















7 Conclusion


Competing Interests



Acknowledgments


References



























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Longitude (degree)

Latit

ude (

degr

ee)

Base station

Terminal point

1km

5km

minus115

minus105

minus100

minus95

minus85

minus75

minus65

minus55

RSL

(dBm

)












radic2 sdot 119865 (]119891 up)100381610038161003816100381610038161003816100381610038161003816) + 11987102

119865 (]119891 up) = int]119891 up

0119890(minus119895sdot120587(11990522))119889119905

(13)




(14)





TX RX

RXTX


RX

TX






radic2 sdot 119865 (]119891 down)100381610038161003816100381610038161003816100381610038161003816)

+ 11989701119865 (]119891 down) = int]119891 down

0119890(minus119895sdot120587(11990522))119889119905

(15)





(17)


120585on (120599) = 10(119871on20)120585down (120599) = 10(119871down20)




Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m6830m

300mhbth =


15m

975m


Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m

6830m

01 02300mhbth =


15m

975m











(19)



01 02

Inland riverLand

Sea level

Tower height

Distance

TX

RX


hb = 2400mhTX

=7955

m

6830m

Wb = 298m

300mhbth =

351164mdt2ob1 = dob2_2r

34mhRX =

15m

975m


5 6 7 8 9 10 114Distance (km)

minus202468

101214161820

Diff

ract

ion

loss

(dB)


RX

TX

Obs

tacle

2

Hei

ght

Ground1756m

401m

Hei

ght157

m

7955

m

161m

700m

877m

Obs

tacle

1



4 5 6 7 8 9 103Distance (km)

0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)


0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)

10 11 12 13 14 159Distance (km)




Scattered rays




065

07

075

08

085

09

095

1

Mag

nitu

de o

f spe

cula

r re

flect

ion

coeffi

cien

t

10 15 20 25 30 35 40 455Distance (km)




Mean plane

IN








w0 = 0003

w0 = 0008

w0 = 002

383895km010203040506070809

1

Shad

owin

g co

effici

ent

05 10 15 20 25 30 35 40 450

Distance (km)




Diveff

= 1


(21)




(22)




5 10 15 20 25 30 35 40 450Distance (km)

0010203040506070809

1

Diff

ract

ion

loss

(dB)

383895km





(23)


1198711 = 20 lg( 1198731radic56561205871205771)

1198712 = 20 lg (1198732)

RE

hTX

dTXhdRXh

h RX

Dis_1Dis_2

Dis_3




minus 088061198713 = 0008612057733 + 0206312057723 + 1109971205773

minus 08934(24)



119871119890 diff = 1198711 + 1198712 minus 100381610038161003816100381611987131003816100381610038161003816 (25)


119871119890 diff = 0 (26)



119871119890 diff = 1198711 + 1198712 + 100381610038161003816100381611987131003816100381610038161003816 (27)




minus35

minus30

minus25

minus20

minus15

minus10

minus5

0

Diff

ract

ion

loss

(dB)

5 10 15 20 25 30 35 40 450Distance (km)


D06 = 113590km

dsum = 383895km



60

80

100

120

140

160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)






110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)


Reflection effect

IR model w20 = 0002

IR model w20 = 0004

IR model w20 = 0008


0 = 002


120572119904 = 120572sub minus 20 lg( 1205824120587) 120573119904 = 120573sub + 2

(29)




120573119897 = 120573urb + 2

(30)


119872119897 = 119872119904 (31)




110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)


Reflection effect

2 3 4 5 6 7 8 9 10 111Distance (km)


IR model h = 05

IR model h = 025

h = 02


4 6 8 10 12 142Distance (km)

60

80

100

120

140

160

180

Prop

agat

ion

path

loss

(dB)


Reflection effect






60708090

100110120130140150160

Prop

agat

ion

path

loss

(dB)

55 6 65 7 755Distance (km)





















7 Conclusion


Competing Interests



Acknowledgments


References



























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation












TX RX

RXTX


RX

TX






radic2 sdot 119865 (]119891 down)100381610038161003816100381610038161003816100381610038161003816)

+ 11989701119865 (]119891 down) = int]119891 down

0119890(minus119895sdot120587(11990522))119889119905

(15)





(17)


120585on (120599) = 10(119871on20)120585down (120599) = 10(119871down20)




Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m6830m

300mhbth =


15m

975m


Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m

6830m

01 02300mhbth =


15m

975m











(19)



01 02

Inland riverLand

Sea level

Tower height

Distance

TX

RX


hb = 2400mhTX

=7955

m

6830m

Wb = 298m

300mhbth =

351164mdt2ob1 = dob2_2r

34mhRX =

15m

975m


5 6 7 8 9 10 114Distance (km)

minus202468

101214161820

Diff

ract

ion

loss

(dB)


RX

TX

Obs

tacle

2

Hei

ght

Ground1756m

401m

Hei

ght157

m

7955

m

161m

700m

877m

Obs

tacle

1



4 5 6 7 8 9 103Distance (km)

0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)


0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)

10 11 12 13 14 159Distance (km)




Scattered rays




065

07

075

08

085

09

095

1

Mag

nitu

de o

f spe

cula

r re

flect

ion

coeffi

cien

t

10 15 20 25 30 35 40 455Distance (km)




Mean plane

IN








w0 = 0003

w0 = 0008

w0 = 002

383895km010203040506070809

1

Shad

owin

g co

effici

ent

05 10 15 20 25 30 35 40 450

Distance (km)




Diveff

= 1


(21)




(22)




5 10 15 20 25 30 35 40 450Distance (km)

0010203040506070809

1

Diff

ract

ion

loss

(dB)

383895km





(23)


1198711 = 20 lg( 1198731radic56561205871205771)

1198712 = 20 lg (1198732)

RE

hTX

dTXhdRXh

h RX

Dis_1Dis_2

Dis_3




minus 088061198713 = 0008612057733 + 0206312057723 + 1109971205773

minus 08934(24)



119871119890 diff = 1198711 + 1198712 minus 100381610038161003816100381611987131003816100381610038161003816 (25)


119871119890 diff = 0 (26)



119871119890 diff = 1198711 + 1198712 + 100381610038161003816100381611987131003816100381610038161003816 (27)




minus35

minus30

minus25

minus20

minus15

minus10

minus5

0

Diff

ract

ion

loss

(dB)

5 10 15 20 25 30 35 40 450Distance (km)


D06 = 113590km

dsum = 383895km



60

80

100

120

140

160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)






110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)


Reflection effect

IR model w20 = 0002

IR model w20 = 0004

IR model w20 = 0008


0 = 002


120572119904 = 120572sub minus 20 lg( 1205824120587) 120573119904 = 120573sub + 2

(29)




120573119897 = 120573urb + 2

(30)


119872119897 = 119872119904 (31)




110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)


Reflection effect

2 3 4 5 6 7 8 9 10 111Distance (km)


IR model h = 05

IR model h = 025

h = 02


4 6 8 10 12 142Distance (km)

60

80

100

120

140

160

180

Prop

agat

ion

path

loss

(dB)


Reflection effect






60708090

100110120130140150160

Prop

agat

ion

path

loss

(dB)

55 6 65 7 755Distance (km)





















7 Conclusion


Competing Interests



Acknowledgments


References



























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m6830m

300mhbth =


15m

975m


Inland river

Land

Sea level

Tower height

Distance

TX

RX


hb = 2400m

Wb = 298m

hTX

=7955

m

6830m

01 02300mhbth =


15m

975m











(19)



01 02

Inland riverLand

Sea level

Tower height

Distance

TX

RX


hb = 2400mhTX

=7955

m

6830m

Wb = 298m

300mhbth =

351164mdt2ob1 = dob2_2r

34mhRX =

15m

975m


5 6 7 8 9 10 114Distance (km)

minus202468

101214161820

Diff

ract

ion

loss

(dB)


RX

TX

Obs

tacle

2

Hei

ght

Ground1756m

401m

Hei

ght157

m

7955

m

161m

700m

877m

Obs

tacle

1



4 5 6 7 8 9 103Distance (km)

0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)


0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)

10 11 12 13 14 159Distance (km)




Scattered rays




065

07

075

08

085

09

095

1

Mag

nitu

de o

f spe

cula

r re

flect

ion

coeffi

cien

t

10 15 20 25 30 35 40 455Distance (km)




Mean plane

IN








w0 = 0003

w0 = 0008

w0 = 002

383895km010203040506070809

1

Shad

owin

g co

effici

ent

05 10 15 20 25 30 35 40 450

Distance (km)




Diveff

= 1


(21)




(22)




5 10 15 20 25 30 35 40 450Distance (km)

0010203040506070809

1

Diff

ract

ion

loss

(dB)

383895km





(23)


1198711 = 20 lg( 1198731radic56561205871205771)

1198712 = 20 lg (1198732)

RE

hTX

dTXhdRXh

h RX

Dis_1Dis_2

Dis_3




minus 088061198713 = 0008612057733 + 0206312057723 + 1109971205773

minus 08934(24)



119871119890 diff = 1198711 + 1198712 minus 100381610038161003816100381611987131003816100381610038161003816 (25)


119871119890 diff = 0 (26)



119871119890 diff = 1198711 + 1198712 + 100381610038161003816100381611987131003816100381610038161003816 (27)




minus35

minus30

minus25

minus20

minus15

minus10

minus5

0

Diff

ract

ion

loss

(dB)

5 10 15 20 25 30 35 40 450Distance (km)


D06 = 113590km

dsum = 383895km



60

80

100

120

140

160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)






110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)


Reflection effect

IR model w20 = 0002

IR model w20 = 0004

IR model w20 = 0008


0 = 002


120572119904 = 120572sub minus 20 lg( 1205824120587) 120573119904 = 120573sub + 2

(29)




120573119897 = 120573urb + 2

(30)


119872119897 = 119872119904 (31)




110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)


Reflection effect

2 3 4 5 6 7 8 9 10 111Distance (km)


IR model h = 05

IR model h = 025

h = 02


4 6 8 10 12 142Distance (km)

60

80

100

120

140

160

180

Prop

agat

ion

path

loss

(dB)


Reflection effect






60708090

100110120130140150160

Prop

agat

ion

path

loss

(dB)

55 6 65 7 755Distance (km)





















7 Conclusion


Competing Interests



Acknowledgments


References



























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










01 02

Inland riverLand

Sea level

Tower height

Distance

TX

RX


hb = 2400mhTX

=7955

m

6830m

Wb = 298m

300mhbth =

351164mdt2ob1 = dob2_2r

34mhRX =

15m

975m


5 6 7 8 9 10 114Distance (km)

minus202468

101214161820

Diff

ract

ion

loss

(dB)


RX

TX

Obs

tacle

2

Hei

ght

Ground1756m

401m

Hei

ght157

m

7955

m

161m

700m

877m

Obs

tacle

1



4 5 6 7 8 9 103Distance (km)

0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)


0

5

10

15

20

25

30

Fini

te d

imen

sions

bui

ldin

gsdi

ffrac

tion

loss

(dB)

10 11 12 13 14 159Distance (km)




Scattered rays




065

07

075

08

085

09

095

1

Mag

nitu

de o

f spe

cula

r re

flect

ion

coeffi

cien

t

10 15 20 25 30 35 40 455Distance (km)




Mean plane

IN








w0 = 0003

w0 = 0008

w0 = 002

383895km010203040506070809

1

Shad

owin

g co

effici

ent

05 10 15 20 25 30 35 40 450

Distance (km)




Diveff

= 1


(21)




(22)




5 10 15 20 25 30 35 40 450Distance (km)

0010203040506070809

1

Diff

ract

ion

loss

(dB)

383895km





(23)


1198711 = 20 lg( 1198731radic56561205871205771)

1198712 = 20 lg (1198732)

RE

hTX

dTXhdRXh

h RX

Dis_1Dis_2

Dis_3




minus 088061198713 = 0008612057733 + 0206312057723 + 1109971205773

minus 08934(24)



119871119890 diff = 1198711 + 1198712 minus 100381610038161003816100381611987131003816100381610038161003816 (25)


119871119890 diff = 0 (26)



119871119890 diff = 1198711 + 1198712 + 100381610038161003816100381611987131003816100381610038161003816 (27)




minus35

minus30

minus25

minus20

minus15

minus10

minus5

0

Diff

ract

ion

loss

(dB)

5 10 15 20 25 30 35 40 450Distance (km)


D06 = 113590km

dsum = 383895km



60

80

100

120

140

160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)






110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)


Reflection effect

IR model w20 = 0002

IR model w20 = 0004

IR model w20 = 0008


0 = 002


120572119904 = 120572sub minus 20 lg( 1205824120587) 120573119904 = 120573sub + 2

(29)




120573119897 = 120573urb + 2

(30)


119872119897 = 119872119904 (31)




110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)


Reflection effect

2 3 4 5 6 7 8 9 10 111Distance (km)


IR model h = 05

IR model h = 025

h = 02


4 6 8 10 12 142Distance (km)

60

80

100

120

140

160

180

Prop

agat

ion

path

loss

(dB)


Reflection effect






60708090

100110120130140150160

Prop

agat

ion

path

loss

(dB)

55 6 65 7 755Distance (km)





















7 Conclusion


Competing Interests



Acknowledgments


References



























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











065

07

075

08

085

09

095

1

Mag

nitu

de o

f spe

cula

r re

flect

ion

coeffi

cien

t

10 15 20 25 30 35 40 455Distance (km)




Mean plane

IN








w0 = 0003

w0 = 0008

w0 = 002

383895km010203040506070809

1

Shad

owin

g co

effici

ent

05 10 15 20 25 30 35 40 450

Distance (km)




Diveff

= 1


(21)




(22)




5 10 15 20 25 30 35 40 450Distance (km)

0010203040506070809

1

Diff

ract

ion

loss

(dB)

383895km





(23)


1198711 = 20 lg( 1198731radic56561205871205771)

1198712 = 20 lg (1198732)

RE

hTX

dTXhdRXh

h RX

Dis_1Dis_2

Dis_3




minus 088061198713 = 0008612057733 + 0206312057723 + 1109971205773

minus 08934(24)



119871119890 diff = 1198711 + 1198712 minus 100381610038161003816100381611987131003816100381610038161003816 (25)


119871119890 diff = 0 (26)



119871119890 diff = 1198711 + 1198712 + 100381610038161003816100381611987131003816100381610038161003816 (27)




minus35

minus30

minus25

minus20

minus15

minus10

minus5

0

Diff

ract

ion

loss

(dB)

5 10 15 20 25 30 35 40 450Distance (km)


D06 = 113590km

dsum = 383895km



60

80

100

120

140

160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)






110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)


Reflection effect

IR model w20 = 0002

IR model w20 = 0004

IR model w20 = 0008


0 = 002


120572119904 = 120572sub minus 20 lg( 1205824120587) 120573119904 = 120573sub + 2

(29)




120573119897 = 120573urb + 2

(30)


119872119897 = 119872119904 (31)




110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)


Reflection effect

2 3 4 5 6 7 8 9 10 111Distance (km)


IR model h = 05

IR model h = 025

h = 02


4 6 8 10 12 142Distance (km)

60

80

100

120

140

160

180

Prop

agat

ion

path

loss

(dB)


Reflection effect






60708090

100110120130140150160

Prop

agat

ion

path

loss

(dB)

55 6 65 7 755Distance (km)





















7 Conclusion


Competing Interests



Acknowledgments


References



























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











5 10 15 20 25 30 35 40 450Distance (km)

0010203040506070809

1

Diff

ract

ion

loss

(dB)

383895km





(23)


1198711 = 20 lg( 1198731radic56561205871205771)

1198712 = 20 lg (1198732)

RE

hTX

dTXhdRXh

h RX

Dis_1Dis_2

Dis_3




minus 088061198713 = 0008612057733 + 0206312057723 + 1109971205773

minus 08934(24)



119871119890 diff = 1198711 + 1198712 minus 100381610038161003816100381611987131003816100381610038161003816 (25)


119871119890 diff = 0 (26)



119871119890 diff = 1198711 + 1198712 + 100381610038161003816100381611987131003816100381610038161003816 (27)




minus35

minus30

minus25

minus20

minus15

minus10

minus5

0

Diff

ract

ion

loss

(dB)

5 10 15 20 25 30 35 40 450Distance (km)


D06 = 113590km

dsum = 383895km



60

80

100

120

140

160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)






110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)


Reflection effect

IR model w20 = 0002

IR model w20 = 0004

IR model w20 = 0008


0 = 002


120572119904 = 120572sub minus 20 lg( 1205824120587) 120573119904 = 120573sub + 2

(29)




120573119897 = 120573urb + 2

(30)


119872119897 = 119872119904 (31)




110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)


Reflection effect

2 3 4 5 6 7 8 9 10 111Distance (km)


IR model h = 05

IR model h = 025

h = 02


4 6 8 10 12 142Distance (km)

60

80

100

120

140

160

180

Prop

agat

ion

path

loss

(dB)


Reflection effect






60708090

100110120130140150160

Prop

agat

ion

path

loss

(dB)

55 6 65 7 755Distance (km)





















7 Conclusion


Competing Interests



Acknowledgments


References



























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










minus35

minus30

minus25

minus20

minus15

minus10

minus5

0

Diff

ract

ion

loss

(dB)

5 10 15 20 25 30 35 40 450Distance (km)


D06 = 113590km

dsum = 383895km



60

80

100

120

140

160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)






110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)

2 3 4 5 6 7 8 9 10 111Distance (km)


Reflection effect

IR model w20 = 0002

IR model w20 = 0004

IR model w20 = 0008


0 = 002


120572119904 = 120572sub minus 20 lg( 1205824120587) 120573119904 = 120573sub + 2

(29)




120573119897 = 120573urb + 2

(30)


119872119897 = 119872119904 (31)




110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)


Reflection effect

2 3 4 5 6 7 8 9 10 111Distance (km)


IR model h = 05

IR model h = 025

h = 02


4 6 8 10 12 142Distance (km)

60

80

100

120

140

160

180

Prop

agat

ion

path

loss

(dB)


Reflection effect






60708090

100110120130140150160

Prop

agat

ion

path

loss

(dB)

55 6 65 7 755Distance (km)





















7 Conclusion


Competing Interests



Acknowledgments


References



























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










110115120125130135140145150155160

Prop

agat

ion

path

loss

(dB)


Reflection effect

2 3 4 5 6 7 8 9 10 111Distance (km)


IR model h = 05

IR model h = 025

h = 02


4 6 8 10 12 142Distance (km)

60

80

100

120

140

160

180

Prop

agat

ion

path

loss

(dB)


Reflection effect






60708090

100110120130140150160

Prop

agat

ion

path

loss

(dB)

55 6 65 7 755Distance (km)





















7 Conclusion


Competing Interests



Acknowledgments


References



























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation




















7 Conclusion


Competing Interests



Acknowledgments


References



























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Acknowledgments


References



























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









Path Loss Channel Model for Inland River Radio...

Documents

Transcript of Path Loss Channel Model for Inland River Radio...