Research ArticleThree-Dimensional Localization Algorithm of WSN NodesBased on RSSI-TOA and Single Mobile Anchor Node

Lieping Zhang Zhenyu Yang Shenglan Zhang and Huanhuan Yang

College of Mechanical and Control Engineering Guilin University of Technology Guilin 541004 China

Correspondence should be addressed to Lieping Zhang zlp_gx_gl163com

Received 20 April 2019 Revised 12 June 2019 Accepted 17 July 2019 Published 28 July 2019

Academic Editor Iickho Song

Copyright copy 2019 Lieping Zhang et al-is is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Aimed at the shortcomings of low localization accuracy of the fixed multianchor method a three-dimensional localizationalgorithm for wireless sensor network nodes is proposed in this paper which combines received signal strength indicator (RSSI)and time of arrival (TOA) ranging information and single mobile anchor node A mobile anchor node was introduced in theproposed three-dimensional localization algorithm for wireless sensor networks firstly and the mobile anchor node movesaccording to the GaussndashMarkov three-dimensional mobility model -en based on the idea of using RSSI ranging in the near endand TOA ranging in the far end a ranging method combining RSSI and TOA ranging information is proposed to obtain theprecise distance between the anchor node and the unknown node Finally the maximum-likelihood estimation method is used toestimate the position of unknown nodes based on the obtained ranging values -e MATLAB simulation results show that theproposed algorithm had a higher localization accuracy and lower localization energy consumption compared with the traditionalRSSI localization method or TOA localization method

1 Introduction

Wireless sensor network (WSN) is a distributed sensornetwork consisting of a large number of inexpensivemicrosensor nodes deployed in the monitoring area to forma multihop and self-organizing network system throughwireless communication [1] One of the basic functions ofthe sensor network is to get the location information of eventoccurrence or the message node However unknown nodesrandomly distributed in the monitoring area cannot locatethemselves in advance so the nodes are necessary to belocated

WSN node localization method can be classified into therange-based localization method and range-free localizationmethod according to whether the distance measurement isneeded in the localization process [2] -e former needs toestimate the distance between the unknown node and theanchor node when estimating the location of unknownnode And the latter estimates the location of unknown nodeby the connectivity of the whole network -erefore theaccuracy of the range-based localization method is better

than that of the range-free localization method Rangingstrategies commonly used in the range-based localizationmethod are angle of arrival (AOA) time of arrival (TOA)and received signal strength indicator (RSSI) [3] In practicalapplications fusion of multiple measurement methods isone of the effective ways to improve the localization effectHan et al [4] proposed a novel indoor positioning algorithmbased on the received signal strength indication and pe-destrian dead reckoning in order to enhance the accuracyand reliability of our proposed probabilistic position se-lection algorithm inmixed line-of-sight (LOS) and non-line-of-sight (NLOS) environments Angelo and Fascista [5] usedthe statistical characterization of the joint maximum-like-lihood estimator to estimate the performance of hybrid RSSIand TOA ranging and proposed a novel closed-form esti-mator based on an ad hoc relaxation of the likelihoodfunction Tomic et al [6] proposed a target location methodby utilizing RSSI and TOA measurements in the adverseNLOS environment

According to whether the anchor node moves or not itcan be divided into static anchor node localization method

HindawiJournal of Electrical and Computer EngineeringVolume 2019 Article ID 4043106 8 pageshttpsdoiorg10115520194043106

and dynamic anchor node localization method [7] Staticanchor node localization requires a certain node densityto meet connectivity requirements so more anchor nodesneed to be deployed -e use of dynamic anchor nodelocalization can greatly reduce the number of anchornodes and can also improve the localization effect Inorder to reduce the number of static anchor nodes reducethe operation cost of the whole network and improve thelocalization effect the localization algorithm based ondynamic anchor nodes has obvious advantages [8] Karimet al [9] proposed a range-free energy-efficient locali-zation technique based on mobile anchor nodes for thelarge-scale machine-to-machine environment Zhao et al[10] proposed an RSSI-based localization algorithmwhich used the RSSI values received by a sensor node frommobile anchor node to estimate the position of the sensornode Singh et al [11] proposed an idea of localizing targetnodes with moving single anchor node using computa-tional intelligence-based application of particle swarmoptimization and H-best particle swarm optimization

In order to improve the ranging accuracy of the node andreduce the localization error of the WSN three-dimensionalnode a three-dimensional localization method of WSNnodes was proposed based on RSSI-TOA and single mobileanchor node through combining the single mobile anchornode strategy and the integrated RSSI-TOA localizationmethod And the effectiveness of the method was verifiedthrough simulation experiments

2 Three-Dimensional Mobile Strategy ofMobile Anchor Node

21 Traditional GaussndashMarkov Mobility Model -e basicprinciple of localization algorithm based on mobile an-chor node is to introduce a mobile anchor node in theWSN monitoring area and let it move according to thepreviously specified model In the process of moving themobile anchor node broadcasts information such as itsposition information periodically and marks the positionin the broadcast which can make full use of the mobilityof the mobile anchor node and form multiple markers-ese markers can be regarded as virtual anchor nodesand their functions are no different from fixed anchornodes -ey can also be used to locate unknown nodesthus reducing the number of fixed anchor nodes requiredby WSN localization [12]

In the localization algorithm of the mobile anchornode its mobility will have a great impact on the local-ization effect and localization energy consumptionRandom waypoint mobility model SCAN DOUBLESCAN and HILBERT mobility models and GaussndashMarkov mobility model are common anchor nodemovement models [13] Among them GaussndashMarkovmobility model is a more realistic mobility model whichhas the characteristics of good flexibility high coverageand strong stability and can cover most of the monitoringareas of WSN [14] It can avoid the situation of smallcoverage probability large blank area and sudden change

of moving trajectory in edge zone In the GaussndashMarkovmobility model the moving speed of the mobile anchornode is a time-dependent GaussndashMarkov process Initialvelocity and direction are given to the mobile anchor nodein the GaussndashMarkov mobility model In a fixed timeperiod T the nodes are moving at a uniform speedaccording to the moving speed After a fixed time periodT the current nodersquos moving speed and direction areupdated -e updated equations are as follows [15]

vt+1 avt +(1minus a)v +1minus a2( )

1113968vn (1)

θt+1 aθt +(1minus a)θ +1minus a2( )

1113968θn (2)

where vt and θt respectively represent the moving speedand direction of the anchor node at t time a is a randomadjustment parameter and has 0le ale 1 When a 0 themobile anchor node is a complete random motion Whena 1 the mobile anchor node becomes a linear motionv and θ respectively represent the moving speed and theaverage of the direction angles of the mobile anchornodes and both are constants As random variables bothvn and θn obey the Gaussian distribution

In the GaussndashMarkov mobility model the position of thenode at the next time period T is determined by the positionvelocity and direction of the previous cycle -e positioncoordinates (xt+1 yt+1) at t + 1 time period T can be ob-tained by the following equation

xt+1 xt + vt middot t middot cos θt (3)

yt+1 yt + vt middot t middot sin θt (4)

22 GaussndashMarkov ree-Dimensional Mobility ModelExtend the traditional GaussndashMarkov mobility model to thethree-dimensional environment When the anchor nodemoves in three-dimensional space there are two angles θt

and βt between the moving direction of velocity and thecoordinate axis

-e angles between the velocity and direction of themobility model and the coordinate axis are shown in thefollowing equations

vt+1 avt +(1minus a)v +1minus a2( )

1113968vn (5)

θt+1 aθt +(1minus a)θ +1minus a2( )

1113968θn (6)

βt+1 aβt +(1minus a)β +1minus a2( )

1113968βn (7)

-e position of the extended GaussndashMarkov mobilitymodel at the next time period T is determined by the po-sition velocity and direction angle of the previous cycleAccording to the moving speed in the moving trajectory theangle of the moving direction and the position of theprevious period the coordinate positions (xt+1 yt+1 zt+1) ofthe next moment can be obtained by equations (8)ndash(10) andthe coordinate position of the mobile anchor node at eachmoment is the position of the virtual anchor node

2 Journal of Electrical and Computer Engineering

xt+1 xt + vt middot cos θt+1 cos βt+1 middot t (8)

xt+1 xt + vt middot cos θt+1 sin βt+1 middot t (9)

zt+1 zt + vt middot cos βt+1 sin θt+1 middot t (10)

In the experiment the WSN location area is regardedas a cube-shaped monitoring area of l length width andheight respectively -e mobile anchor node starts fromthe randomly occurring starting point coordinates(xt yt zt) and moves along the mobility model trajectorywith the velocity vt and communication radius R andperiodically broadcasts its own coordinate informationand ID location information with T as the time intervalAfter completing a broadcast the next movement followsAt this time the mobile anchor node moves at a speed vt+1and the moving direction of the speed and the two axesexisting in the coordinate axis are θt+1 and βt+1 -emovement is repeated continuously so that it can beenabled to cover the entire network When the mobileanchor node moves to the edge of the network the mobileanchor node changes its speed and direction and starts anew movement In the process of moving when theanchor node realizes the localization of the last unknownnode to be measured the mobile anchor node does notmove anymore and stops working A three-dimensionalmoving trajectory of a GaussndashMarkov mobility model isshown in Figure 1

3 RangingPrincipleofFusionRSSIMethodandTOA Method

31 RSSI Ranging Principle RSSI ranging method calculatesthe power loss in the process of signal propagation throughthe signal power received by sensor nodes and then convertsthe power loss into distance value by using theoretical andempirical models But for the relationship between powerloss and distance different signal transmission models havedifferent results In practical application the log-normalmodel is the most widely used signal transmission models[16] -e relationship between signal strength and distanceof the model is shown in the following equation

P P0 minus 10n lgd

d01113888 1113889 + Xσ (11)

where d and d0 are the distances of the receiving node andthe reference node to the signal transmitter P0 and n are thequantities of the channel characteristics P0 is the signalstrength of the reference point and n represents the path lossindex which varies with the environment Xσ is a randomnoise obeying a log-normal distribution with a mean of zeroand a standard deviation of σ Its main function is to describethe influence of uncertain factors such as signal reflectionand noise interference on the received signal strength It canbe expressed by means and variances of multiple signalmeasurements and its distribution density function can beexpressed as follows

f(x) 12π

radicσ

e(xminusμ)2minus2σ2

(12)

where the specific expressions of μ and σ are as follows

μ 1k

1113944

k

i1RSSIi (13)

σ

1kminus 1

1113944

k

i1RSSIi minus μ( 1113857

2

11139741113972

(14)

where RSSIi represents the signal strength value of the ithsignal and k represents the total number of measurements-e empirical value of Xσ can be obtained by setting upreasonable experiments

RSSI ranging method will cause a large measurementerror due to fading at a long distance which means thatwhen the measurement distance increases the ranging errorwill increase significantly resulting in inaccuratelocalization

32 TOARanging Principle TOA ranging method estimatesthe distance between two nodes by measuring the trans-mission time of the wireless signal -e ranging equation isshown as follows

d c middot t (15)

where d and t are the distance values and propagation timesbetween the signal transmitter and the receiver respectivelyand c is the speed of transmission signal

In the localization process the TOA ranging localizationalgorithm needs the nodes of the transmitter and the receiverto transmit signals at the same time which makes the TOAranging method greatly limited in practical applications-e

200

150

100

50

0200

100

0 050

100150

200

Figure 1-ree-dimensionalmoving trajectory of theGaussndashMarkovmobility model

Journal of Electrical and Computer Engineering 3

IEEE 802154a protocol provides the symmetric double-sided two-way ranging (SDS-TWR) method for TOAranging [17] which can offset the impact of the delay timedue to the inability to meet time synchronization on theranging results -e SDS-TWR method uses a symmetricalmethod to measure TOA in two directions -e specificimplementation process is as follows (1) First TWR node Aserves as a signal transmitter and node B serves as a signalreceiver (2) Second TWR node B serves as a signaltransmitter and node A serves as a signal receiver -emeasurement process can be described as shown in Figure 2

In this way a standard time tTOA a standard time that isthe final timemeasurement result can be obtained as shownas follows

tTOA troundA minus treplyB + troundB minus treplyA

4 (16)

When the signal transmitter is close to the receiver theTOA ranging algorithmmeasurement error will have a greatinfluence on the accuracy of the TOA ranging localizationalgorithm resulting in an excessive deviation between themeasured distance value and the actual distance value

33 Ranging Implementation of Fusion RSSI and TOAMethod -rough the analysis of the RSSI and TOA rangingprinciple we know that the path loss values of the RSSIranging method vary greatly due to various factors in dif-ferent environments [18] When measuring at close rangethe signal transmission basically obeys the log normal losslaw used in the RSSI ranging technology and the rangingaccuracy is high When the unknown node is far away fromthe anchor node the transmission power of the signal isattenuated faster and ranging error will be increased On theother hand the TOA ranging method still needs a higherprecision time measuring device in the calculation processand there is a measurement error in the close range mea-surement which will cause a large deviation in the nodelocalization process But when it is measured at a long rangeit has a high precision [19] Aiming at the advantages anddisadvantages of RSSI and TOA ranging methods RSSIranging method is proposed for the proximal end and TOAranging method for the distal end in our paper which iscalled RSSI-TOA ranging method In the RSSI-TOA rangingmethod the distance threshold using RSSI ranging or TOAranging is selected by the experimental method which isdescribed in Section 54 -e specific ranging process isshown in Figure 3

4 Three-Dimensional Location Method forWSN Nodes

On the basis of synthesizing the advantages and disadvan-tages of RSSI and TOA ranging methods the mobile anchornode is introduced and a three-dimensional localizationalgorithm forWSN nodes is proposed which combines RSSIand TOA ranging methods -e maximum-likelihood es-timation method is used to calculate the position of WSNnodes -e implementation steps are as follows

(1) Data information acquisitionN wireless sensor nodes Si(i 1 2 N) are ran-domly deployed in the three-dimensional WSNmonitoring area Q Among these sensor nodes thereare L anchor nodes Bi(i 1 2 L) Nminus L unknownnodes and virtual anchor nodes Bj(j 1 2 L)periodically broadcast by the mobile anchor nodes Itis stipulated that all sensor nodes in the monitoringarea have the same communication radius R

(2) RSSI ranging methodRSSI ranging method is used to measure the distanceof the node and the logarithmic normal wirelesssignal transmission model is used to obtain the at-tenuation value of the signal -en its attenuationvalue is transformed into the corresponding distance

Node A Node B

tTOA

treply B

tround B

tround A

treply A

Figure 2 Measuring process of the TOA ranging method

Start

Known detection area Qand node Si (xi yi zi)

Unknown node receives anchornode signal detects RSSI and

calculates distance value

Distance valuelt distance threshold

N

Y Ranging byTOA

Record RSSIranging values

Finish ranging

Calculating by TOAranging formula and

recording distance value

Figure 3 Ranging process with the RSSI and TOA method

4 Journal of Electrical and Computer Engineering

value di according to equation (11) In the WSNmonitoring area when the distance between theanchor node and the unknown node is less than theprescribed threshold in the communication range ofthe unknown node the distance value is recorded

(3) TOA ranging methodIn the WSN monitoring area when the distancebetween the anchor node and the unknown node isgreater than the prescribed threshold and less thanthe communication radius tTOA the measurementdistance dj from the anchor node to the unknownnode can be obtained according to the time obtainedby equation (16) and the distance value is recorded

(4) Estimation of coordinate positionAt the node localization stage the coordinate valuesof unknown nodes are obtained by maximum-likelihood estimation after the measured distancevalue di(or dj) between m(mge 4) unknown nodesand anchor nodes (or virtual anchor nodes) is ob-tained from the three-dimensional monitoring areaof WSN

5 Experimental Simulation and Analysis

51 Setting of Operating Environment and ParametersLocalization algorithm is simulated in the MATLAB sim-ulation environment -e simulation space is the WSNthree-dimensional space of 200m times 200m times 200m A mo-bile anchor node is introduced in the proposed method -emobile anchor node moves periodically in the WSN mon-itoring area according to the GaussndashMarkov three-di-mensional mobility model and broadcasts its own data andcoordinate position information every a period T to formmultiple virtual anchor nodes In the initialization process200 sensor nodes are randomly generated in three-di-mensional space of which the number of unknown nodes tobe located is 180 and the number of fixed anchor nodes is 20If the location of the node is deployed in an obstructed area

it will be redeployed Once deployed the position of allnodes cannot be changed -e communication radius of allnodes is set to 40m the transmission speed c of the signals isset to light speed the reference node distance d0 is 1m P0 isset to minus30 the path loss index n is 2 the noise Xσ is 3 and themesh spacing t of WSN is 25 In order to be closer to theactual environment an obstacle area is set up in this ex-periment which is a cube of size 20m times 20m times 20m Fixedanchor nodes unknown nodes to be localization and ob-stacle areas of the simulation experiment are shown inFigure 4-e green and red points represent unknown nodesand fixed anchor nodes respectively in Figure 4 All ex-perimental results are performed on 100 experimentalsimulations and then the average of all experimental resultsis used as the final result

52 Influence of Anchor Density on Localization Error with orwithout Mobile Anchor Node In the process of WSN nodelocalization the higher the density of the anchor nodes in thenetwork the easier it is to cover the entire WSN monitoringarea which can effectively reduce the localization errorHowever the density of the anchor node increases whichgenerates a lot of redundant information also increases thevarious burdens of the WSN -erefore it is not that thehigher the density of the anchor nodes the better the lo-calization effect As shown in Figure 5 the average locali-zation errors of RSSI method TOA method and RSSI-TOAmethod under different anchor node densities are discussedin the case of mobile anchor nodes It can be seen fromFigure 5 that when there is a mobile anchor node the av-erage localization errors decrease with the increase of thedensity of the fixed anchor node When the density of thefixed anchor nodes is 10 three algorithms reach a relativelyideal level Even if the fixed anchor node density is increasedthe localization errors of three algorithms no longer changesignificantly It can also be seen that under the same con-ditions the average localization error of the RSSI-TOAmethod is the smallest

0200

2040

180

60

160

80

200

100

140 180

120

120 160

140

140100

160

120

180

80 100

200

60 806040 4020 200 0

Figure 4 Distribution of node initial position

Journal of Electrical and Computer Engineering 5

In Figure 6 there are the average localization errors ofthe three algorithms under different anchor node densitieswithout a mobile anchor node It can be seen from Figure 6that when the anchor node is not moved the density of thefixed anchor node is increased and the average localizationerrors of three methods tend to decrease continuously andthe average localization error of the RSSI-TOAmethod is thesmallest When the density of the fixed anchor node reaches25 the density of the fixed anchor node is increased andthe localization error does not change much At the sametime when the density of the fixed anchor node reaches 30the average localization errors of three localization algo-rithms perform little different from the average localizationerror of the mobile anchor node with a fixed anchor noderatio of about 10 It can be concluded that by introducing amobile anchor node in advance and adopting RSSI-TOAmethod not only the cost of networking and localization andthe power consumption can be reduced but also the error ofnode localization can be reduced in the same time and theperformance of node localization can be improved

53 Influence of Communication Radius on LocalizationError In the WSN monitoring area when the communi-cation radius of the node is large more anchor nodes will becaptured and the corresponding node localization error willbe reduced Figure 7 shows the average localization errors ofthree localization algorithms under different communica-tion radius when amobile anchor node is introduced and thedensity of the fixed anchor node is 10 It can be seen fromFigure 7 that when the node communication radius is in-creased from 10m to 40m the average localization errors ofthree methods are reduced but the localization error of theRSSI-TOA method is the smallest

54 Influence of Measurement Position on Localization ErrorLocalization error is affected by the measurement position ofthe nodes With the increase of measuring position distance

the localization error will increase accordingly Figure 8shows the relationship between the measuring positionand the average localization error when a mobile anchornode is introduced and the density of the fixed anchor nodeis 10 It can be seen from Figure 8 that the localizationerror of the RSSI method is smaller than that of the TOAmethod in the range of 0ndash12m With the increase ofmeasurement distance the localization error of the RSSImethod is larger than that of the TOA method when themeasurement distance is more than 12m -e RSSI-TOAmethod proposed in this paper has the smallest averagelocalization error under the same conditions

55 Localization Error of Unknown Nodes Figure 9 showsthe localization errors of 180 unknown nodes obtained by

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30Anchor node density ()

RSSI methodTOA methodRSSI-TOA method

Figure 5 Relationship between localization error and anchor nodedensity with mobile anchor nodes

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30Anchor node density ()

RSSI methodTOA methodRSSI-TOA method

Figure 6 Relationship between localization error and anchor nodedensity without mobile anchor node

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

RSSI methodTOA methodRSSI-TOA method

10 15 20 25 30 35 40Communication radius (m)

Figure 7 Relationship between communication radius and av-erage localization error

6 Journal of Electrical and Computer Engineering

the three methods It can be seen from Figure 9 that the nodelocalization errors of RSSI method and TOA method fluc-tuate greatly and there are many nodes having large errorsHowever the curve fluctuation of the node localization errorof the RSSI-TOAmethod proposed in this paper is relativelysmall and the localization error is the smallest among thethree methods It shows that in the three-dimensional en-vironment the RSSI-TOA method proposed in this paper is

the least affected by the external environment and has thehighest localization accuracy and the best localizationstability

6 Conclusion

In this work a ranging method is proposed that combinesRSSI and TOA ranging information in this paper -e

151617181920

1413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30 35 40Measuring position (m)

RSSI methodTOA methodRSSI-TOA method

Figure 8 Relationship between localization error and measurement position

RSSI methodTOA methodRSSI-TOA method

25

20

15

10

5

0

Aver

age l

ocal

izat

ion

erro

r (m

)

0 20 40 60 80 100 120 140 160 180Node number

Figure 9 Localization error of each node in three localization algorithms

Journal of Electrical and Computer Engineering 7

proposed ranging method reduces the ranging errors byusing RSSI or TOA ranging step by step in the ranging stageIn order to further improve the accuracy of node localiza-tion a mobile anchor node is introduced-e anchor node ismoved in the three-dimensional environment by GaussndashMarkov to form virtual anchor nodes -en the maximum-likelihood estimation method is used to realize the three-dimensional localization of WSN nodes -e simulationresults verified the effectiveness of the proposed RSSI-TOAmethod and achieved better localization results Because theRSSI ranging value and the TOA ranging value are greatlyaffected by WSN node devices and external environmentthe difference of the different node devices in differentapplication environments at different times will be relativelylarge Research on LSSVR node localization algorithm thatcombines RSSI and TOA ranging information in differentenvironments will be our future research At the same timethe static nodes are located in the three-dimensionalmonitoring area through amobile anchor node in this paper-e next step is to study the method of localization of themobile node by multiple mobile anchor nodes

Data Availability

Data sharing is not applicable to this article as no datasetswere generated or analysed during the current study

Conflicts of Interest

-e authors declare that they have no conflicts of interest

Acknowledgments

-is work was supported by the National Natural ScienceFoundation of China (no 61741303) Guangxi NaturalScience Foundation (no 2017GXNSFAA198161) KeyLaboratory of Spatial Information and Geomatics (GuilinUniversity of Technology) (nos 15-140-07-23 and 16-380-25-23) and Innovation Project of Guangxi Graduate Edu-cation (noYCSW2019159)

References

[1] R E Mohamed A I Saleh M Abdelrazzak and A S SamraldquoSurvey on wireless sensor network applications and energyefficient routing protocolsrdquo Wireless Personal Communica-tions vol 101 no 2 pp 1019ndash1055 2018

[2] K Langendoen and N Reijers ldquoDistributed localization inwireless sensor networksa quantitative comparisonrdquo Com-puter Networks vol 43 no 4 pp 499ndash518 2013

[3] G Han J Jiang C Zhang T Q Duong M Guizani andG K Karagiannidis ldquoA survey on mobile anchor nodeassisted localization in wireless sensor networksrdquo IEEECommunications Surveys amp Tutorials vol 18 no 3pp 2220ndash2243 2016

[4] K Han H Xing Z Deng and Y Du ldquoA RSSIPDR-basedprobabilistic position selection algorithm with NLOS iden-tification for indoor localisationrdquo ISPRS International Journalof Geo-Information vol 7 no 6 pp 232ndash253 2018

[5] C Angelo and A Fascista ldquoOn the hybrid TOARSS rangeestimation in wireless sensor networksrdquo IEEE Transactions onWireless Communications vol 17 no 1 pp 361ndash371 2018

[6] S Tomic M Beko M Tuba and V M F Correia ldquoTargetlocalization in NLOS environments using RSS and TOAmeasurementsrdquo IEEE Wireless Communications Lettersvol 7 no 6 pp 1062ndash1065 2018

[7] E Erdemir and T E Tuncer ldquoPath planning for mobile-anchor based wireless sensor network localization static anddynamic schemesrdquo Ad Hoc Networks vol 77 pp 1ndash10 2018

[8] Y Chen S Lu J Chen and T Ren ldquoNode localization al-gorithm of wireless sensor networks with mobile beaconnoderdquo Peer-to-Peer Networking and Applications vol 10no 3 pp 1ndash13 2016

[9] L Karim N Nasser Q H Mahmoud A Anpalagan andT E Salti ldquoRange-free localization approach for M2Mcommunication system using mobile anchor nodesrdquo Journalof Network and Computer Applications vol 47 pp 137ndash1462015

[10] Y Zhao J Xu and J Jiang ldquoRSSI based localization withmobile anchor for wireless sensor networksrdquo in Proceedingsof the 2017 International Conference on Geo-Spatial Knowledgeand Intelligence pp 176ndash187 Chiang Mai -ailand De-cember 2018

[11] P Singh A Khosla A Kumar and M Khosla ldquoOptimizedlocalization of target nodes using single mobile anchor node inwireless sensor networkrdquo AEUmdashInternational Journal ofElectronics and Communications vol 91 pp 55ndash65 2018

[12] P Singh A Khosla A Kumar and M Khosla ldquoComputa-tional intelligence based localization of moving target nodesusing single anchor node in wireless sensor networksrdquoTelecommunication Systems vol 69 no 3 pp 397ndash411 2018

[13] X C Song Y S Zhao and L ZWang ldquoGaussndashMarkov-basedmobile anchor localization (GM-MAL) algorithm based onlocal linear embedding optimization in internet of sensornetworksrdquo Cognitive Systems Research vol 52 pp 138ndash1432018

[14] T Das and S Roy ldquoEnergy efficient and event driven mobilitymodel in mobile WSNrdquo in Proceedings of the IEEE In-ternational Conference on Advanced Networks amp Telecom-munications Systems pp 1ndash6 Kolkata India December 2015

[15] Z Zhong D-Y Luo S-Q Liu X-P Fan and Z-H Qu ldquoAnadaptive localization approach for wireless sensor networksbased on gauss-markov mobility modelrdquo Acta AutomaticaSinica vol 36 no 11 pp 1557ndash1568 2010

[16] V Bianchi P Ciampolini and I De Munari ldquoRSSI-basedindoor localization and identification for zigbee wirelesssensor networks in smart homesrdquo IEEE Transactions on In-strumentation and Measurement vol 68 no 2 pp 566ndash5752019

[17] F Despaux K Jaffres-Runser A V D Bossche and T ValldquoAccurate and platform-agnostic time-of-flight estimation inultra-wide bandrdquo in Proceeding of the 27th IEEE InternationalSymposium on Personal Indoor and Mobile Radio Commu-nications pp 1ndash7 Valencia Spain September 2016

[18] S Sadowski and P Spachos ldquoRSSI-based indoor localizationwith the Internet of thingsrdquo IEEEAccess vol 6 pp 30149ndash301612018

[19] S Wu S Zhang and D Huang ldquoA TOA-based localizationalgorithm with simultaneous NLOS mitigation and syn-chronization error eliminationrdquo IEEE Sensors Letters vol 3no 3 pp 1ndash4 2019

8 Journal of Electrical and Computer Engineering

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

xt+1 xt + vt middot cos θt+1 cos βt+1 middot t (8)

xt+1 xt + vt middot cos θt+1 sin βt+1 middot t (9)

zt+1 zt + vt middot cos βt+1 sin θt+1 middot t (10)

In the experiment the WSN location area is regardedas a cube-shaped monitoring area of l length width andheight respectively -e mobile anchor node starts fromthe randomly occurring starting point coordinates(xt yt zt) and moves along the mobility model trajectorywith the velocity vt and communication radius R andperiodically broadcasts its own coordinate informationand ID location information with T as the time intervalAfter completing a broadcast the next movement followsAt this time the mobile anchor node moves at a speed vt+1and the moving direction of the speed and the two axesexisting in the coordinate axis are θt+1 and βt+1 -emovement is repeated continuously so that it can beenabled to cover the entire network When the mobileanchor node moves to the edge of the network the mobileanchor node changes its speed and direction and starts anew movement In the process of moving when theanchor node realizes the localization of the last unknownnode to be measured the mobile anchor node does notmove anymore and stops working A three-dimensionalmoving trajectory of a GaussndashMarkov mobility model isshown in Figure 1

3 RangingPrincipleofFusionRSSIMethodandTOA Method

31 RSSI Ranging Principle RSSI ranging method calculatesthe power loss in the process of signal propagation throughthe signal power received by sensor nodes and then convertsthe power loss into distance value by using theoretical andempirical models But for the relationship between powerloss and distance different signal transmission models havedifferent results In practical application the log-normalmodel is the most widely used signal transmission models[16] -e relationship between signal strength and distanceof the model is shown in the following equation

P P0 minus 10n lgd

d01113888 1113889 + Xσ (11)

where d and d0 are the distances of the receiving node andthe reference node to the signal transmitter P0 and n are thequantities of the channel characteristics P0 is the signalstrength of the reference point and n represents the path lossindex which varies with the environment Xσ is a randomnoise obeying a log-normal distribution with a mean of zeroand a standard deviation of σ Its main function is to describethe influence of uncertain factors such as signal reflectionand noise interference on the received signal strength It canbe expressed by means and variances of multiple signalmeasurements and its distribution density function can beexpressed as follows

f(x) 12π

radicσ

e(xminusμ)2minus2σ2

(12)

where the specific expressions of μ and σ are as follows

μ 1k

1113944

k

i1RSSIi (13)

σ

1kminus 1

1113944

k

i1RSSIi minus μ( 1113857

2

11139741113972

(14)

where RSSIi represents the signal strength value of the ithsignal and k represents the total number of measurements-e empirical value of Xσ can be obtained by setting upreasonable experiments

RSSI ranging method will cause a large measurementerror due to fading at a long distance which means thatwhen the measurement distance increases the ranging errorwill increase significantly resulting in inaccuratelocalization

32 TOARanging Principle TOA ranging method estimatesthe distance between two nodes by measuring the trans-mission time of the wireless signal -e ranging equation isshown as follows

d c middot t (15)

where d and t are the distance values and propagation timesbetween the signal transmitter and the receiver respectivelyand c is the speed of transmission signal

In the localization process the TOA ranging localizationalgorithm needs the nodes of the transmitter and the receiverto transmit signals at the same time which makes the TOAranging method greatly limited in practical applications-e

200

150

100

50

0200

100

0 050

100150

200

Figure 1-ree-dimensionalmoving trajectory of theGaussndashMarkovmobility model

Journal of Electrical and Computer Engineering 3

IEEE 802154a protocol provides the symmetric double-sided two-way ranging (SDS-TWR) method for TOAranging [17] which can offset the impact of the delay timedue to the inability to meet time synchronization on theranging results -e SDS-TWR method uses a symmetricalmethod to measure TOA in two directions -e specificimplementation process is as follows (1) First TWR node Aserves as a signal transmitter and node B serves as a signalreceiver (2) Second TWR node B serves as a signaltransmitter and node A serves as a signal receiver -emeasurement process can be described as shown in Figure 2

In this way a standard time tTOA a standard time that isthe final timemeasurement result can be obtained as shownas follows

tTOA troundA minus treplyB + troundB minus treplyA

4 (16)

When the signal transmitter is close to the receiver theTOA ranging algorithmmeasurement error will have a greatinfluence on the accuracy of the TOA ranging localizationalgorithm resulting in an excessive deviation between themeasured distance value and the actual distance value

33 Ranging Implementation of Fusion RSSI and TOAMethod -rough the analysis of the RSSI and TOA rangingprinciple we know that the path loss values of the RSSIranging method vary greatly due to various factors in dif-ferent environments [18] When measuring at close rangethe signal transmission basically obeys the log normal losslaw used in the RSSI ranging technology and the rangingaccuracy is high When the unknown node is far away fromthe anchor node the transmission power of the signal isattenuated faster and ranging error will be increased On theother hand the TOA ranging method still needs a higherprecision time measuring device in the calculation processand there is a measurement error in the close range mea-surement which will cause a large deviation in the nodelocalization process But when it is measured at a long rangeit has a high precision [19] Aiming at the advantages anddisadvantages of RSSI and TOA ranging methods RSSIranging method is proposed for the proximal end and TOAranging method for the distal end in our paper which iscalled RSSI-TOA ranging method In the RSSI-TOA rangingmethod the distance threshold using RSSI ranging or TOAranging is selected by the experimental method which isdescribed in Section 54 -e specific ranging process isshown in Figure 3

4 Three-Dimensional Location Method forWSN Nodes

On the basis of synthesizing the advantages and disadvan-tages of RSSI and TOA ranging methods the mobile anchornode is introduced and a three-dimensional localizationalgorithm forWSN nodes is proposed which combines RSSIand TOA ranging methods -e maximum-likelihood es-timation method is used to calculate the position of WSNnodes -e implementation steps are as follows

(1) Data information acquisitionN wireless sensor nodes Si(i 1 2 N) are ran-domly deployed in the three-dimensional WSNmonitoring area Q Among these sensor nodes thereare L anchor nodes Bi(i 1 2 L) Nminus L unknownnodes and virtual anchor nodes Bj(j 1 2 L)periodically broadcast by the mobile anchor nodes Itis stipulated that all sensor nodes in the monitoringarea have the same communication radius R

(2) RSSI ranging methodRSSI ranging method is used to measure the distanceof the node and the logarithmic normal wirelesssignal transmission model is used to obtain the at-tenuation value of the signal -en its attenuationvalue is transformed into the corresponding distance

Node A Node B

tTOA

treply B

tround B

tround A

treply A

Figure 2 Measuring process of the TOA ranging method

Start

Known detection area Qand node Si (xi yi zi)

Unknown node receives anchornode signal detects RSSI and

calculates distance value

Distance valuelt distance threshold

N

Y Ranging byTOA

Record RSSIranging values

Finish ranging

Calculating by TOAranging formula and

recording distance value

Figure 3 Ranging process with the RSSI and TOA method

4 Journal of Electrical and Computer Engineering

value di according to equation (11) In the WSNmonitoring area when the distance between theanchor node and the unknown node is less than theprescribed threshold in the communication range ofthe unknown node the distance value is recorded

(3) TOA ranging methodIn the WSN monitoring area when the distancebetween the anchor node and the unknown node isgreater than the prescribed threshold and less thanthe communication radius tTOA the measurementdistance dj from the anchor node to the unknownnode can be obtained according to the time obtainedby equation (16) and the distance value is recorded

(4) Estimation of coordinate positionAt the node localization stage the coordinate valuesof unknown nodes are obtained by maximum-likelihood estimation after the measured distancevalue di(or dj) between m(mge 4) unknown nodesand anchor nodes (or virtual anchor nodes) is ob-tained from the three-dimensional monitoring areaof WSN

5 Experimental Simulation and Analysis

51 Setting of Operating Environment and ParametersLocalization algorithm is simulated in the MATLAB sim-ulation environment -e simulation space is the WSNthree-dimensional space of 200m times 200m times 200m A mo-bile anchor node is introduced in the proposed method -emobile anchor node moves periodically in the WSN mon-itoring area according to the GaussndashMarkov three-di-mensional mobility model and broadcasts its own data andcoordinate position information every a period T to formmultiple virtual anchor nodes In the initialization process200 sensor nodes are randomly generated in three-di-mensional space of which the number of unknown nodes tobe located is 180 and the number of fixed anchor nodes is 20If the location of the node is deployed in an obstructed area

it will be redeployed Once deployed the position of allnodes cannot be changed -e communication radius of allnodes is set to 40m the transmission speed c of the signals isset to light speed the reference node distance d0 is 1m P0 isset to minus30 the path loss index n is 2 the noise Xσ is 3 and themesh spacing t of WSN is 25 In order to be closer to theactual environment an obstacle area is set up in this ex-periment which is a cube of size 20m times 20m times 20m Fixedanchor nodes unknown nodes to be localization and ob-stacle areas of the simulation experiment are shown inFigure 4-e green and red points represent unknown nodesand fixed anchor nodes respectively in Figure 4 All ex-perimental results are performed on 100 experimentalsimulations and then the average of all experimental resultsis used as the final result

52 Influence of Anchor Density on Localization Error with orwithout Mobile Anchor Node In the process of WSN nodelocalization the higher the density of the anchor nodes in thenetwork the easier it is to cover the entire WSN monitoringarea which can effectively reduce the localization errorHowever the density of the anchor node increases whichgenerates a lot of redundant information also increases thevarious burdens of the WSN -erefore it is not that thehigher the density of the anchor nodes the better the lo-calization effect As shown in Figure 5 the average locali-zation errors of RSSI method TOA method and RSSI-TOAmethod under different anchor node densities are discussedin the case of mobile anchor nodes It can be seen fromFigure 5 that when there is a mobile anchor node the av-erage localization errors decrease with the increase of thedensity of the fixed anchor node When the density of thefixed anchor nodes is 10 three algorithms reach a relativelyideal level Even if the fixed anchor node density is increasedthe localization errors of three algorithms no longer changesignificantly It can also be seen that under the same con-ditions the average localization error of the RSSI-TOAmethod is the smallest

0200

2040

180

60

160

80

200

100

140 180

120

120 160

140

140100

160

120

180

80 100

200

60 806040 4020 200 0

Figure 4 Distribution of node initial position

Journal of Electrical and Computer Engineering 5

In Figure 6 there are the average localization errors ofthe three algorithms under different anchor node densitieswithout a mobile anchor node It can be seen from Figure 6that when the anchor node is not moved the density of thefixed anchor node is increased and the average localizationerrors of three methods tend to decrease continuously andthe average localization error of the RSSI-TOAmethod is thesmallest When the density of the fixed anchor node reaches25 the density of the fixed anchor node is increased andthe localization error does not change much At the sametime when the density of the fixed anchor node reaches 30the average localization errors of three localization algo-rithms perform little different from the average localizationerror of the mobile anchor node with a fixed anchor noderatio of about 10 It can be concluded that by introducing amobile anchor node in advance and adopting RSSI-TOAmethod not only the cost of networking and localization andthe power consumption can be reduced but also the error ofnode localization can be reduced in the same time and theperformance of node localization can be improved

53 Influence of Communication Radius on LocalizationError In the WSN monitoring area when the communi-cation radius of the node is large more anchor nodes will becaptured and the corresponding node localization error willbe reduced Figure 7 shows the average localization errors ofthree localization algorithms under different communica-tion radius when amobile anchor node is introduced and thedensity of the fixed anchor node is 10 It can be seen fromFigure 7 that when the node communication radius is in-creased from 10m to 40m the average localization errors ofthree methods are reduced but the localization error of theRSSI-TOA method is the smallest

54 Influence of Measurement Position on Localization ErrorLocalization error is affected by the measurement position ofthe nodes With the increase of measuring position distance

the localization error will increase accordingly Figure 8shows the relationship between the measuring positionand the average localization error when a mobile anchornode is introduced and the density of the fixed anchor nodeis 10 It can be seen from Figure 8 that the localizationerror of the RSSI method is smaller than that of the TOAmethod in the range of 0ndash12m With the increase ofmeasurement distance the localization error of the RSSImethod is larger than that of the TOA method when themeasurement distance is more than 12m -e RSSI-TOAmethod proposed in this paper has the smallest averagelocalization error under the same conditions

55 Localization Error of Unknown Nodes Figure 9 showsthe localization errors of 180 unknown nodes obtained by

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30Anchor node density ()

RSSI methodTOA methodRSSI-TOA method

Figure 5 Relationship between localization error and anchor nodedensity with mobile anchor nodes

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30Anchor node density ()

RSSI methodTOA methodRSSI-TOA method

Figure 6 Relationship between localization error and anchor nodedensity without mobile anchor node

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

RSSI methodTOA methodRSSI-TOA method

10 15 20 25 30 35 40Communication radius (m)

Figure 7 Relationship between communication radius and av-erage localization error

6 Journal of Electrical and Computer Engineering

the three methods It can be seen from Figure 9 that the nodelocalization errors of RSSI method and TOA method fluc-tuate greatly and there are many nodes having large errorsHowever the curve fluctuation of the node localization errorof the RSSI-TOAmethod proposed in this paper is relativelysmall and the localization error is the smallest among thethree methods It shows that in the three-dimensional en-vironment the RSSI-TOA method proposed in this paper is

the least affected by the external environment and has thehighest localization accuracy and the best localizationstability

6 Conclusion

In this work a ranging method is proposed that combinesRSSI and TOA ranging information in this paper -e

151617181920

1413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30 35 40Measuring position (m)

RSSI methodTOA methodRSSI-TOA method

Figure 8 Relationship between localization error and measurement position

RSSI methodTOA methodRSSI-TOA method

25

20

15

10

5

0

Aver

age l

ocal

izat

ion

erro

r (m

)

0 20 40 60 80 100 120 140 160 180Node number

Figure 9 Localization error of each node in three localization algorithms

Journal of Electrical and Computer Engineering 7

proposed ranging method reduces the ranging errors byusing RSSI or TOA ranging step by step in the ranging stageIn order to further improve the accuracy of node localiza-tion a mobile anchor node is introduced-e anchor node ismoved in the three-dimensional environment by GaussndashMarkov to form virtual anchor nodes -en the maximum-likelihood estimation method is used to realize the three-dimensional localization of WSN nodes -e simulationresults verified the effectiveness of the proposed RSSI-TOAmethod and achieved better localization results Because theRSSI ranging value and the TOA ranging value are greatlyaffected by WSN node devices and external environmentthe difference of the different node devices in differentapplication environments at different times will be relativelylarge Research on LSSVR node localization algorithm thatcombines RSSI and TOA ranging information in differentenvironments will be our future research At the same timethe static nodes are located in the three-dimensionalmonitoring area through amobile anchor node in this paper-e next step is to study the method of localization of themobile node by multiple mobile anchor nodes

Data Availability

Data sharing is not applicable to this article as no datasetswere generated or analysed during the current study

Conflicts of Interest

-e authors declare that they have no conflicts of interest

Acknowledgments

-is work was supported by the National Natural ScienceFoundation of China (no 61741303) Guangxi NaturalScience Foundation (no 2017GXNSFAA198161) KeyLaboratory of Spatial Information and Geomatics (GuilinUniversity of Technology) (nos 15-140-07-23 and 16-380-25-23) and Innovation Project of Guangxi Graduate Edu-cation (noYCSW2019159)

References

[1] R E Mohamed A I Saleh M Abdelrazzak and A S SamraldquoSurvey on wireless sensor network applications and energyefficient routing protocolsrdquo Wireless Personal Communica-tions vol 101 no 2 pp 1019ndash1055 2018

[2] K Langendoen and N Reijers ldquoDistributed localization inwireless sensor networksa quantitative comparisonrdquo Com-puter Networks vol 43 no 4 pp 499ndash518 2013

[3] G Han J Jiang C Zhang T Q Duong M Guizani andG K Karagiannidis ldquoA survey on mobile anchor nodeassisted localization in wireless sensor networksrdquo IEEECommunications Surveys amp Tutorials vol 18 no 3pp 2220ndash2243 2016

[4] K Han H Xing Z Deng and Y Du ldquoA RSSIPDR-basedprobabilistic position selection algorithm with NLOS iden-tification for indoor localisationrdquo ISPRS International Journalof Geo-Information vol 7 no 6 pp 232ndash253 2018

[5] C Angelo and A Fascista ldquoOn the hybrid TOARSS rangeestimation in wireless sensor networksrdquo IEEE Transactions onWireless Communications vol 17 no 1 pp 361ndash371 2018

[6] S Tomic M Beko M Tuba and V M F Correia ldquoTargetlocalization in NLOS environments using RSS and TOAmeasurementsrdquo IEEE Wireless Communications Lettersvol 7 no 6 pp 1062ndash1065 2018

[7] E Erdemir and T E Tuncer ldquoPath planning for mobile-anchor based wireless sensor network localization static anddynamic schemesrdquo Ad Hoc Networks vol 77 pp 1ndash10 2018

[8] Y Chen S Lu J Chen and T Ren ldquoNode localization al-gorithm of wireless sensor networks with mobile beaconnoderdquo Peer-to-Peer Networking and Applications vol 10no 3 pp 1ndash13 2016

[9] L Karim N Nasser Q H Mahmoud A Anpalagan andT E Salti ldquoRange-free localization approach for M2Mcommunication system using mobile anchor nodesrdquo Journalof Network and Computer Applications vol 47 pp 137ndash1462015

[10] Y Zhao J Xu and J Jiang ldquoRSSI based localization withmobile anchor for wireless sensor networksrdquo in Proceedingsof the 2017 International Conference on Geo-Spatial Knowledgeand Intelligence pp 176ndash187 Chiang Mai -ailand De-cember 2018

[11] P Singh A Khosla A Kumar and M Khosla ldquoOptimizedlocalization of target nodes using single mobile anchor node inwireless sensor networkrdquo AEUmdashInternational Journal ofElectronics and Communications vol 91 pp 55ndash65 2018

[12] P Singh A Khosla A Kumar and M Khosla ldquoComputa-tional intelligence based localization of moving target nodesusing single anchor node in wireless sensor networksrdquoTelecommunication Systems vol 69 no 3 pp 397ndash411 2018

[13] X C Song Y S Zhao and L ZWang ldquoGaussndashMarkov-basedmobile anchor localization (GM-MAL) algorithm based onlocal linear embedding optimization in internet of sensornetworksrdquo Cognitive Systems Research vol 52 pp 138ndash1432018

[14] T Das and S Roy ldquoEnergy efficient and event driven mobilitymodel in mobile WSNrdquo in Proceedings of the IEEE In-ternational Conference on Advanced Networks amp Telecom-munications Systems pp 1ndash6 Kolkata India December 2015

[15] Z Zhong D-Y Luo S-Q Liu X-P Fan and Z-H Qu ldquoAnadaptive localization approach for wireless sensor networksbased on gauss-markov mobility modelrdquo Acta AutomaticaSinica vol 36 no 11 pp 1557ndash1568 2010

[16] V Bianchi P Ciampolini and I De Munari ldquoRSSI-basedindoor localization and identification for zigbee wirelesssensor networks in smart homesrdquo IEEE Transactions on In-strumentation and Measurement vol 68 no 2 pp 566ndash5752019

[17] F Despaux K Jaffres-Runser A V D Bossche and T ValldquoAccurate and platform-agnostic time-of-flight estimation inultra-wide bandrdquo in Proceeding of the 27th IEEE InternationalSymposium on Personal Indoor and Mobile Radio Commu-nications pp 1ndash7 Valencia Spain September 2016

[18] S Sadowski and P Spachos ldquoRSSI-based indoor localizationwith the Internet of thingsrdquo IEEEAccess vol 6 pp 30149ndash301612018

[19] S Wu S Zhang and D Huang ldquoA TOA-based localizationalgorithm with simultaneous NLOS mitigation and syn-chronization error eliminationrdquo IEEE Sensors Letters vol 3no 3 pp 1ndash4 2019

8 Journal of Electrical and Computer Engineering

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

IEEE 802154a protocol provides the symmetric double-sided two-way ranging (SDS-TWR) method for TOAranging [17] which can offset the impact of the delay timedue to the inability to meet time synchronization on theranging results -e SDS-TWR method uses a symmetricalmethod to measure TOA in two directions -e specificimplementation process is as follows (1) First TWR node Aserves as a signal transmitter and node B serves as a signalreceiver (2) Second TWR node B serves as a signaltransmitter and node A serves as a signal receiver -emeasurement process can be described as shown in Figure 2

In this way a standard time tTOA a standard time that isthe final timemeasurement result can be obtained as shownas follows

tTOA troundA minus treplyB + troundB minus treplyA

4 (16)

When the signal transmitter is close to the receiver theTOA ranging algorithmmeasurement error will have a greatinfluence on the accuracy of the TOA ranging localizationalgorithm resulting in an excessive deviation between themeasured distance value and the actual distance value

33 Ranging Implementation of Fusion RSSI and TOAMethod -rough the analysis of the RSSI and TOA rangingprinciple we know that the path loss values of the RSSIranging method vary greatly due to various factors in dif-ferent environments [18] When measuring at close rangethe signal transmission basically obeys the log normal losslaw used in the RSSI ranging technology and the rangingaccuracy is high When the unknown node is far away fromthe anchor node the transmission power of the signal isattenuated faster and ranging error will be increased On theother hand the TOA ranging method still needs a higherprecision time measuring device in the calculation processand there is a measurement error in the close range mea-surement which will cause a large deviation in the nodelocalization process But when it is measured at a long rangeit has a high precision [19] Aiming at the advantages anddisadvantages of RSSI and TOA ranging methods RSSIranging method is proposed for the proximal end and TOAranging method for the distal end in our paper which iscalled RSSI-TOA ranging method In the RSSI-TOA rangingmethod the distance threshold using RSSI ranging or TOAranging is selected by the experimental method which isdescribed in Section 54 -e specific ranging process isshown in Figure 3

4 Three-Dimensional Location Method forWSN Nodes

On the basis of synthesizing the advantages and disadvan-tages of RSSI and TOA ranging methods the mobile anchornode is introduced and a three-dimensional localizationalgorithm forWSN nodes is proposed which combines RSSIand TOA ranging methods -e maximum-likelihood es-timation method is used to calculate the position of WSNnodes -e implementation steps are as follows

(1) Data information acquisitionN wireless sensor nodes Si(i 1 2 N) are ran-domly deployed in the three-dimensional WSNmonitoring area Q Among these sensor nodes thereare L anchor nodes Bi(i 1 2 L) Nminus L unknownnodes and virtual anchor nodes Bj(j 1 2 L)periodically broadcast by the mobile anchor nodes Itis stipulated that all sensor nodes in the monitoringarea have the same communication radius R

(2) RSSI ranging methodRSSI ranging method is used to measure the distanceof the node and the logarithmic normal wirelesssignal transmission model is used to obtain the at-tenuation value of the signal -en its attenuationvalue is transformed into the corresponding distance

Node A Node B

tTOA

treply B

tround B

tround A

treply A

Figure 2 Measuring process of the TOA ranging method

Start

Known detection area Qand node Si (xi yi zi)

Unknown node receives anchornode signal detects RSSI and

calculates distance value

Distance valuelt distance threshold

N

Y Ranging byTOA

Record RSSIranging values

Finish ranging

Calculating by TOAranging formula and

recording distance value

Figure 3 Ranging process with the RSSI and TOA method

4 Journal of Electrical and Computer Engineering

value di according to equation (11) In the WSNmonitoring area when the distance between theanchor node and the unknown node is less than theprescribed threshold in the communication range ofthe unknown node the distance value is recorded

(3) TOA ranging methodIn the WSN monitoring area when the distancebetween the anchor node and the unknown node isgreater than the prescribed threshold and less thanthe communication radius tTOA the measurementdistance dj from the anchor node to the unknownnode can be obtained according to the time obtainedby equation (16) and the distance value is recorded

(4) Estimation of coordinate positionAt the node localization stage the coordinate valuesof unknown nodes are obtained by maximum-likelihood estimation after the measured distancevalue di(or dj) between m(mge 4) unknown nodesand anchor nodes (or virtual anchor nodes) is ob-tained from the three-dimensional monitoring areaof WSN

5 Experimental Simulation and Analysis

51 Setting of Operating Environment and ParametersLocalization algorithm is simulated in the MATLAB sim-ulation environment -e simulation space is the WSNthree-dimensional space of 200m times 200m times 200m A mo-bile anchor node is introduced in the proposed method -emobile anchor node moves periodically in the WSN mon-itoring area according to the GaussndashMarkov three-di-mensional mobility model and broadcasts its own data andcoordinate position information every a period T to formmultiple virtual anchor nodes In the initialization process200 sensor nodes are randomly generated in three-di-mensional space of which the number of unknown nodes tobe located is 180 and the number of fixed anchor nodes is 20If the location of the node is deployed in an obstructed area

it will be redeployed Once deployed the position of allnodes cannot be changed -e communication radius of allnodes is set to 40m the transmission speed c of the signals isset to light speed the reference node distance d0 is 1m P0 isset to minus30 the path loss index n is 2 the noise Xσ is 3 and themesh spacing t of WSN is 25 In order to be closer to theactual environment an obstacle area is set up in this ex-periment which is a cube of size 20m times 20m times 20m Fixedanchor nodes unknown nodes to be localization and ob-stacle areas of the simulation experiment are shown inFigure 4-e green and red points represent unknown nodesand fixed anchor nodes respectively in Figure 4 All ex-perimental results are performed on 100 experimentalsimulations and then the average of all experimental resultsis used as the final result

52 Influence of Anchor Density on Localization Error with orwithout Mobile Anchor Node In the process of WSN nodelocalization the higher the density of the anchor nodes in thenetwork the easier it is to cover the entire WSN monitoringarea which can effectively reduce the localization errorHowever the density of the anchor node increases whichgenerates a lot of redundant information also increases thevarious burdens of the WSN -erefore it is not that thehigher the density of the anchor nodes the better the lo-calization effect As shown in Figure 5 the average locali-zation errors of RSSI method TOA method and RSSI-TOAmethod under different anchor node densities are discussedin the case of mobile anchor nodes It can be seen fromFigure 5 that when there is a mobile anchor node the av-erage localization errors decrease with the increase of thedensity of the fixed anchor node When the density of thefixed anchor nodes is 10 three algorithms reach a relativelyideal level Even if the fixed anchor node density is increasedthe localization errors of three algorithms no longer changesignificantly It can also be seen that under the same con-ditions the average localization error of the RSSI-TOAmethod is the smallest

0200

2040

180

60

160

80

200

100

140 180

120

120 160

140

140100

160

120

180

80 100

200

60 806040 4020 200 0

Figure 4 Distribution of node initial position

Journal of Electrical and Computer Engineering 5

In Figure 6 there are the average localization errors ofthe three algorithms under different anchor node densitieswithout a mobile anchor node It can be seen from Figure 6that when the anchor node is not moved the density of thefixed anchor node is increased and the average localizationerrors of three methods tend to decrease continuously andthe average localization error of the RSSI-TOAmethod is thesmallest When the density of the fixed anchor node reaches25 the density of the fixed anchor node is increased andthe localization error does not change much At the sametime when the density of the fixed anchor node reaches 30the average localization errors of three localization algo-rithms perform little different from the average localizationerror of the mobile anchor node with a fixed anchor noderatio of about 10 It can be concluded that by introducing amobile anchor node in advance and adopting RSSI-TOAmethod not only the cost of networking and localization andthe power consumption can be reduced but also the error ofnode localization can be reduced in the same time and theperformance of node localization can be improved

53 Influence of Communication Radius on LocalizationError In the WSN monitoring area when the communi-cation radius of the node is large more anchor nodes will becaptured and the corresponding node localization error willbe reduced Figure 7 shows the average localization errors ofthree localization algorithms under different communica-tion radius when amobile anchor node is introduced and thedensity of the fixed anchor node is 10 It can be seen fromFigure 7 that when the node communication radius is in-creased from 10m to 40m the average localization errors ofthree methods are reduced but the localization error of theRSSI-TOA method is the smallest

54 Influence of Measurement Position on Localization ErrorLocalization error is affected by the measurement position ofthe nodes With the increase of measuring position distance

the localization error will increase accordingly Figure 8shows the relationship between the measuring positionand the average localization error when a mobile anchornode is introduced and the density of the fixed anchor nodeis 10 It can be seen from Figure 8 that the localizationerror of the RSSI method is smaller than that of the TOAmethod in the range of 0ndash12m With the increase ofmeasurement distance the localization error of the RSSImethod is larger than that of the TOA method when themeasurement distance is more than 12m -e RSSI-TOAmethod proposed in this paper has the smallest averagelocalization error under the same conditions

55 Localization Error of Unknown Nodes Figure 9 showsthe localization errors of 180 unknown nodes obtained by

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30Anchor node density ()

RSSI methodTOA methodRSSI-TOA method

Figure 5 Relationship between localization error and anchor nodedensity with mobile anchor nodes

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30Anchor node density ()

RSSI methodTOA methodRSSI-TOA method

Figure 6 Relationship between localization error and anchor nodedensity without mobile anchor node

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

RSSI methodTOA methodRSSI-TOA method

10 15 20 25 30 35 40Communication radius (m)

Figure 7 Relationship between communication radius and av-erage localization error

6 Journal of Electrical and Computer Engineering

the three methods It can be seen from Figure 9 that the nodelocalization errors of RSSI method and TOA method fluc-tuate greatly and there are many nodes having large errorsHowever the curve fluctuation of the node localization errorof the RSSI-TOAmethod proposed in this paper is relativelysmall and the localization error is the smallest among thethree methods It shows that in the three-dimensional en-vironment the RSSI-TOA method proposed in this paper is

the least affected by the external environment and has thehighest localization accuracy and the best localizationstability

6 Conclusion

In this work a ranging method is proposed that combinesRSSI and TOA ranging information in this paper -e

151617181920

1413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30 35 40Measuring position (m)

RSSI methodTOA methodRSSI-TOA method

Figure 8 Relationship between localization error and measurement position

RSSI methodTOA methodRSSI-TOA method

25

20

15

10

5

0

Aver

age l

ocal

izat

ion

erro

r (m

)

0 20 40 60 80 100 120 140 160 180Node number

Figure 9 Localization error of each node in three localization algorithms

Journal of Electrical and Computer Engineering 7

proposed ranging method reduces the ranging errors byusing RSSI or TOA ranging step by step in the ranging stageIn order to further improve the accuracy of node localiza-tion a mobile anchor node is introduced-e anchor node ismoved in the three-dimensional environment by GaussndashMarkov to form virtual anchor nodes -en the maximum-likelihood estimation method is used to realize the three-dimensional localization of WSN nodes -e simulationresults verified the effectiveness of the proposed RSSI-TOAmethod and achieved better localization results Because theRSSI ranging value and the TOA ranging value are greatlyaffected by WSN node devices and external environmentthe difference of the different node devices in differentapplication environments at different times will be relativelylarge Research on LSSVR node localization algorithm thatcombines RSSI and TOA ranging information in differentenvironments will be our future research At the same timethe static nodes are located in the three-dimensionalmonitoring area through amobile anchor node in this paper-e next step is to study the method of localization of themobile node by multiple mobile anchor nodes

Data Availability

Data sharing is not applicable to this article as no datasetswere generated or analysed during the current study

Conflicts of Interest

-e authors declare that they have no conflicts of interest

Acknowledgments

-is work was supported by the National Natural ScienceFoundation of China (no 61741303) Guangxi NaturalScience Foundation (no 2017GXNSFAA198161) KeyLaboratory of Spatial Information and Geomatics (GuilinUniversity of Technology) (nos 15-140-07-23 and 16-380-25-23) and Innovation Project of Guangxi Graduate Edu-cation (noYCSW2019159)

References

[1] R E Mohamed A I Saleh M Abdelrazzak and A S SamraldquoSurvey on wireless sensor network applications and energyefficient routing protocolsrdquo Wireless Personal Communica-tions vol 101 no 2 pp 1019ndash1055 2018

[2] K Langendoen and N Reijers ldquoDistributed localization inwireless sensor networksa quantitative comparisonrdquo Com-puter Networks vol 43 no 4 pp 499ndash518 2013

[3] G Han J Jiang C Zhang T Q Duong M Guizani andG K Karagiannidis ldquoA survey on mobile anchor nodeassisted localization in wireless sensor networksrdquo IEEECommunications Surveys amp Tutorials vol 18 no 3pp 2220ndash2243 2016

[4] K Han H Xing Z Deng and Y Du ldquoA RSSIPDR-basedprobabilistic position selection algorithm with NLOS iden-tification for indoor localisationrdquo ISPRS International Journalof Geo-Information vol 7 no 6 pp 232ndash253 2018

[5] C Angelo and A Fascista ldquoOn the hybrid TOARSS rangeestimation in wireless sensor networksrdquo IEEE Transactions onWireless Communications vol 17 no 1 pp 361ndash371 2018

[6] S Tomic M Beko M Tuba and V M F Correia ldquoTargetlocalization in NLOS environments using RSS and TOAmeasurementsrdquo IEEE Wireless Communications Lettersvol 7 no 6 pp 1062ndash1065 2018

[7] E Erdemir and T E Tuncer ldquoPath planning for mobile-anchor based wireless sensor network localization static anddynamic schemesrdquo Ad Hoc Networks vol 77 pp 1ndash10 2018

[8] Y Chen S Lu J Chen and T Ren ldquoNode localization al-gorithm of wireless sensor networks with mobile beaconnoderdquo Peer-to-Peer Networking and Applications vol 10no 3 pp 1ndash13 2016

[9] L Karim N Nasser Q H Mahmoud A Anpalagan andT E Salti ldquoRange-free localization approach for M2Mcommunication system using mobile anchor nodesrdquo Journalof Network and Computer Applications vol 47 pp 137ndash1462015

[10] Y Zhao J Xu and J Jiang ldquoRSSI based localization withmobile anchor for wireless sensor networksrdquo in Proceedingsof the 2017 International Conference on Geo-Spatial Knowledgeand Intelligence pp 176ndash187 Chiang Mai -ailand De-cember 2018

[11] P Singh A Khosla A Kumar and M Khosla ldquoOptimizedlocalization of target nodes using single mobile anchor node inwireless sensor networkrdquo AEUmdashInternational Journal ofElectronics and Communications vol 91 pp 55ndash65 2018

[12] P Singh A Khosla A Kumar and M Khosla ldquoComputa-tional intelligence based localization of moving target nodesusing single anchor node in wireless sensor networksrdquoTelecommunication Systems vol 69 no 3 pp 397ndash411 2018

[13] X C Song Y S Zhao and L ZWang ldquoGaussndashMarkov-basedmobile anchor localization (GM-MAL) algorithm based onlocal linear embedding optimization in internet of sensornetworksrdquo Cognitive Systems Research vol 52 pp 138ndash1432018

[14] T Das and S Roy ldquoEnergy efficient and event driven mobilitymodel in mobile WSNrdquo in Proceedings of the IEEE In-ternational Conference on Advanced Networks amp Telecom-munications Systems pp 1ndash6 Kolkata India December 2015

[15] Z Zhong D-Y Luo S-Q Liu X-P Fan and Z-H Qu ldquoAnadaptive localization approach for wireless sensor networksbased on gauss-markov mobility modelrdquo Acta AutomaticaSinica vol 36 no 11 pp 1557ndash1568 2010

[16] V Bianchi P Ciampolini and I De Munari ldquoRSSI-basedindoor localization and identification for zigbee wirelesssensor networks in smart homesrdquo IEEE Transactions on In-strumentation and Measurement vol 68 no 2 pp 566ndash5752019

[17] F Despaux K Jaffres-Runser A V D Bossche and T ValldquoAccurate and platform-agnostic time-of-flight estimation inultra-wide bandrdquo in Proceeding of the 27th IEEE InternationalSymposium on Personal Indoor and Mobile Radio Commu-nications pp 1ndash7 Valencia Spain September 2016

[18] S Sadowski and P Spachos ldquoRSSI-based indoor localizationwith the Internet of thingsrdquo IEEEAccess vol 6 pp 30149ndash301612018

[19] S Wu S Zhang and D Huang ldquoA TOA-based localizationalgorithm with simultaneous NLOS mitigation and syn-chronization error eliminationrdquo IEEE Sensors Letters vol 3no 3 pp 1ndash4 2019

8 Journal of Electrical and Computer Engineering

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

value di according to equation (11) In the WSNmonitoring area when the distance between theanchor node and the unknown node is less than theprescribed threshold in the communication range ofthe unknown node the distance value is recorded

(3) TOA ranging methodIn the WSN monitoring area when the distancebetween the anchor node and the unknown node isgreater than the prescribed threshold and less thanthe communication radius tTOA the measurementdistance dj from the anchor node to the unknownnode can be obtained according to the time obtainedby equation (16) and the distance value is recorded

(4) Estimation of coordinate positionAt the node localization stage the coordinate valuesof unknown nodes are obtained by maximum-likelihood estimation after the measured distancevalue di(or dj) between m(mge 4) unknown nodesand anchor nodes (or virtual anchor nodes) is ob-tained from the three-dimensional monitoring areaof WSN

5 Experimental Simulation and Analysis

51 Setting of Operating Environment and ParametersLocalization algorithm is simulated in the MATLAB sim-ulation environment -e simulation space is the WSNthree-dimensional space of 200m times 200m times 200m A mo-bile anchor node is introduced in the proposed method -emobile anchor node moves periodically in the WSN mon-itoring area according to the GaussndashMarkov three-di-mensional mobility model and broadcasts its own data andcoordinate position information every a period T to formmultiple virtual anchor nodes In the initialization process200 sensor nodes are randomly generated in three-di-mensional space of which the number of unknown nodes tobe located is 180 and the number of fixed anchor nodes is 20If the location of the node is deployed in an obstructed area

it will be redeployed Once deployed the position of allnodes cannot be changed -e communication radius of allnodes is set to 40m the transmission speed c of the signals isset to light speed the reference node distance d0 is 1m P0 isset to minus30 the path loss index n is 2 the noise Xσ is 3 and themesh spacing t of WSN is 25 In order to be closer to theactual environment an obstacle area is set up in this ex-periment which is a cube of size 20m times 20m times 20m Fixedanchor nodes unknown nodes to be localization and ob-stacle areas of the simulation experiment are shown inFigure 4-e green and red points represent unknown nodesand fixed anchor nodes respectively in Figure 4 All ex-perimental results are performed on 100 experimentalsimulations and then the average of all experimental resultsis used as the final result

52 Influence of Anchor Density on Localization Error with orwithout Mobile Anchor Node In the process of WSN nodelocalization the higher the density of the anchor nodes in thenetwork the easier it is to cover the entire WSN monitoringarea which can effectively reduce the localization errorHowever the density of the anchor node increases whichgenerates a lot of redundant information also increases thevarious burdens of the WSN -erefore it is not that thehigher the density of the anchor nodes the better the lo-calization effect As shown in Figure 5 the average locali-zation errors of RSSI method TOA method and RSSI-TOAmethod under different anchor node densities are discussedin the case of mobile anchor nodes It can be seen fromFigure 5 that when there is a mobile anchor node the av-erage localization errors decrease with the increase of thedensity of the fixed anchor node When the density of thefixed anchor nodes is 10 three algorithms reach a relativelyideal level Even if the fixed anchor node density is increasedthe localization errors of three algorithms no longer changesignificantly It can also be seen that under the same con-ditions the average localization error of the RSSI-TOAmethod is the smallest

0200

2040

180

60

160

80

200

100

140 180

120

120 160

140

140100

160

120

180

80 100

200

60 806040 4020 200 0

Figure 4 Distribution of node initial position

Journal of Electrical and Computer Engineering 5

In Figure 6 there are the average localization errors ofthe three algorithms under different anchor node densitieswithout a mobile anchor node It can be seen from Figure 6that when the anchor node is not moved the density of thefixed anchor node is increased and the average localizationerrors of three methods tend to decrease continuously andthe average localization error of the RSSI-TOAmethod is thesmallest When the density of the fixed anchor node reaches25 the density of the fixed anchor node is increased andthe localization error does not change much At the sametime when the density of the fixed anchor node reaches 30the average localization errors of three localization algo-rithms perform little different from the average localizationerror of the mobile anchor node with a fixed anchor noderatio of about 10 It can be concluded that by introducing amobile anchor node in advance and adopting RSSI-TOAmethod not only the cost of networking and localization andthe power consumption can be reduced but also the error ofnode localization can be reduced in the same time and theperformance of node localization can be improved

53 Influence of Communication Radius on LocalizationError In the WSN monitoring area when the communi-cation radius of the node is large more anchor nodes will becaptured and the corresponding node localization error willbe reduced Figure 7 shows the average localization errors ofthree localization algorithms under different communica-tion radius when amobile anchor node is introduced and thedensity of the fixed anchor node is 10 It can be seen fromFigure 7 that when the node communication radius is in-creased from 10m to 40m the average localization errors ofthree methods are reduced but the localization error of theRSSI-TOA method is the smallest

54 Influence of Measurement Position on Localization ErrorLocalization error is affected by the measurement position ofthe nodes With the increase of measuring position distance

the localization error will increase accordingly Figure 8shows the relationship between the measuring positionand the average localization error when a mobile anchornode is introduced and the density of the fixed anchor nodeis 10 It can be seen from Figure 8 that the localizationerror of the RSSI method is smaller than that of the TOAmethod in the range of 0ndash12m With the increase ofmeasurement distance the localization error of the RSSImethod is larger than that of the TOA method when themeasurement distance is more than 12m -e RSSI-TOAmethod proposed in this paper has the smallest averagelocalization error under the same conditions

55 Localization Error of Unknown Nodes Figure 9 showsthe localization errors of 180 unknown nodes obtained by

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30Anchor node density ()

RSSI methodTOA methodRSSI-TOA method

Figure 5 Relationship between localization error and anchor nodedensity with mobile anchor nodes

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30Anchor node density ()

RSSI methodTOA methodRSSI-TOA method

Figure 6 Relationship between localization error and anchor nodedensity without mobile anchor node

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

RSSI methodTOA methodRSSI-TOA method

10 15 20 25 30 35 40Communication radius (m)

Figure 7 Relationship between communication radius and av-erage localization error

6 Journal of Electrical and Computer Engineering

the three methods It can be seen from Figure 9 that the nodelocalization errors of RSSI method and TOA method fluc-tuate greatly and there are many nodes having large errorsHowever the curve fluctuation of the node localization errorof the RSSI-TOAmethod proposed in this paper is relativelysmall and the localization error is the smallest among thethree methods It shows that in the three-dimensional en-vironment the RSSI-TOA method proposed in this paper is

the least affected by the external environment and has thehighest localization accuracy and the best localizationstability

6 Conclusion

In this work a ranging method is proposed that combinesRSSI and TOA ranging information in this paper -e

151617181920

1413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30 35 40Measuring position (m)

RSSI methodTOA methodRSSI-TOA method

Figure 8 Relationship between localization error and measurement position

RSSI methodTOA methodRSSI-TOA method

25

20

15

10

5

0

Aver

age l

ocal

izat

ion

erro

r (m

)

0 20 40 60 80 100 120 140 160 180Node number

Figure 9 Localization error of each node in three localization algorithms

Journal of Electrical and Computer Engineering 7

proposed ranging method reduces the ranging errors byusing RSSI or TOA ranging step by step in the ranging stageIn order to further improve the accuracy of node localiza-tion a mobile anchor node is introduced-e anchor node ismoved in the three-dimensional environment by GaussndashMarkov to form virtual anchor nodes -en the maximum-likelihood estimation method is used to realize the three-dimensional localization of WSN nodes -e simulationresults verified the effectiveness of the proposed RSSI-TOAmethod and achieved better localization results Because theRSSI ranging value and the TOA ranging value are greatlyaffected by WSN node devices and external environmentthe difference of the different node devices in differentapplication environments at different times will be relativelylarge Research on LSSVR node localization algorithm thatcombines RSSI and TOA ranging information in differentenvironments will be our future research At the same timethe static nodes are located in the three-dimensionalmonitoring area through amobile anchor node in this paper-e next step is to study the method of localization of themobile node by multiple mobile anchor nodes

Data Availability

Data sharing is not applicable to this article as no datasetswere generated or analysed during the current study

Conflicts of Interest

-e authors declare that they have no conflicts of interest

Acknowledgments

-is work was supported by the National Natural ScienceFoundation of China (no 61741303) Guangxi NaturalScience Foundation (no 2017GXNSFAA198161) KeyLaboratory of Spatial Information and Geomatics (GuilinUniversity of Technology) (nos 15-140-07-23 and 16-380-25-23) and Innovation Project of Guangxi Graduate Edu-cation (noYCSW2019159)

References

[1] R E Mohamed A I Saleh M Abdelrazzak and A S SamraldquoSurvey on wireless sensor network applications and energyefficient routing protocolsrdquo Wireless Personal Communica-tions vol 101 no 2 pp 1019ndash1055 2018

[2] K Langendoen and N Reijers ldquoDistributed localization inwireless sensor networksa quantitative comparisonrdquo Com-puter Networks vol 43 no 4 pp 499ndash518 2013

[3] G Han J Jiang C Zhang T Q Duong M Guizani andG K Karagiannidis ldquoA survey on mobile anchor nodeassisted localization in wireless sensor networksrdquo IEEECommunications Surveys amp Tutorials vol 18 no 3pp 2220ndash2243 2016

[4] K Han H Xing Z Deng and Y Du ldquoA RSSIPDR-basedprobabilistic position selection algorithm with NLOS iden-tification for indoor localisationrdquo ISPRS International Journalof Geo-Information vol 7 no 6 pp 232ndash253 2018

[5] C Angelo and A Fascista ldquoOn the hybrid TOARSS rangeestimation in wireless sensor networksrdquo IEEE Transactions onWireless Communications vol 17 no 1 pp 361ndash371 2018

[6] S Tomic M Beko M Tuba and V M F Correia ldquoTargetlocalization in NLOS environments using RSS and TOAmeasurementsrdquo IEEE Wireless Communications Lettersvol 7 no 6 pp 1062ndash1065 2018

[7] E Erdemir and T E Tuncer ldquoPath planning for mobile-anchor based wireless sensor network localization static anddynamic schemesrdquo Ad Hoc Networks vol 77 pp 1ndash10 2018

[8] Y Chen S Lu J Chen and T Ren ldquoNode localization al-gorithm of wireless sensor networks with mobile beaconnoderdquo Peer-to-Peer Networking and Applications vol 10no 3 pp 1ndash13 2016

[9] L Karim N Nasser Q H Mahmoud A Anpalagan andT E Salti ldquoRange-free localization approach for M2Mcommunication system using mobile anchor nodesrdquo Journalof Network and Computer Applications vol 47 pp 137ndash1462015

[10] Y Zhao J Xu and J Jiang ldquoRSSI based localization withmobile anchor for wireless sensor networksrdquo in Proceedingsof the 2017 International Conference on Geo-Spatial Knowledgeand Intelligence pp 176ndash187 Chiang Mai -ailand De-cember 2018

[11] P Singh A Khosla A Kumar and M Khosla ldquoOptimizedlocalization of target nodes using single mobile anchor node inwireless sensor networkrdquo AEUmdashInternational Journal ofElectronics and Communications vol 91 pp 55ndash65 2018

[12] P Singh A Khosla A Kumar and M Khosla ldquoComputa-tional intelligence based localization of moving target nodesusing single anchor node in wireless sensor networksrdquoTelecommunication Systems vol 69 no 3 pp 397ndash411 2018

[13] X C Song Y S Zhao and L ZWang ldquoGaussndashMarkov-basedmobile anchor localization (GM-MAL) algorithm based onlocal linear embedding optimization in internet of sensornetworksrdquo Cognitive Systems Research vol 52 pp 138ndash1432018

[14] T Das and S Roy ldquoEnergy efficient and event driven mobilitymodel in mobile WSNrdquo in Proceedings of the IEEE In-ternational Conference on Advanced Networks amp Telecom-munications Systems pp 1ndash6 Kolkata India December 2015

[15] Z Zhong D-Y Luo S-Q Liu X-P Fan and Z-H Qu ldquoAnadaptive localization approach for wireless sensor networksbased on gauss-markov mobility modelrdquo Acta AutomaticaSinica vol 36 no 11 pp 1557ndash1568 2010

[16] V Bianchi P Ciampolini and I De Munari ldquoRSSI-basedindoor localization and identification for zigbee wirelesssensor networks in smart homesrdquo IEEE Transactions on In-strumentation and Measurement vol 68 no 2 pp 566ndash5752019

[17] F Despaux K Jaffres-Runser A V D Bossche and T ValldquoAccurate and platform-agnostic time-of-flight estimation inultra-wide bandrdquo in Proceeding of the 27th IEEE InternationalSymposium on Personal Indoor and Mobile Radio Commu-nications pp 1ndash7 Valencia Spain September 2016

[18] S Sadowski and P Spachos ldquoRSSI-based indoor localizationwith the Internet of thingsrdquo IEEEAccess vol 6 pp 30149ndash301612018

[19] S Wu S Zhang and D Huang ldquoA TOA-based localizationalgorithm with simultaneous NLOS mitigation and syn-chronization error eliminationrdquo IEEE Sensors Letters vol 3no 3 pp 1ndash4 2019

8 Journal of Electrical and Computer Engineering

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

In Figure 6 there are the average localization errors ofthe three algorithms under different anchor node densitieswithout a mobile anchor node It can be seen from Figure 6that when the anchor node is not moved the density of thefixed anchor node is increased and the average localizationerrors of three methods tend to decrease continuously andthe average localization error of the RSSI-TOAmethod is thesmallest When the density of the fixed anchor node reaches25 the density of the fixed anchor node is increased andthe localization error does not change much At the sametime when the density of the fixed anchor node reaches 30the average localization errors of three localization algo-rithms perform little different from the average localizationerror of the mobile anchor node with a fixed anchor noderatio of about 10 It can be concluded that by introducing amobile anchor node in advance and adopting RSSI-TOAmethod not only the cost of networking and localization andthe power consumption can be reduced but also the error ofnode localization can be reduced in the same time and theperformance of node localization can be improved

53 Influence of Communication Radius on LocalizationError In the WSN monitoring area when the communi-cation radius of the node is large more anchor nodes will becaptured and the corresponding node localization error willbe reduced Figure 7 shows the average localization errors ofthree localization algorithms under different communica-tion radius when amobile anchor node is introduced and thedensity of the fixed anchor node is 10 It can be seen fromFigure 7 that when the node communication radius is in-creased from 10m to 40m the average localization errors ofthree methods are reduced but the localization error of theRSSI-TOA method is the smallest

54 Influence of Measurement Position on Localization ErrorLocalization error is affected by the measurement position ofthe nodes With the increase of measuring position distance

the localization error will increase accordingly Figure 8shows the relationship between the measuring positionand the average localization error when a mobile anchornode is introduced and the density of the fixed anchor nodeis 10 It can be seen from Figure 8 that the localizationerror of the RSSI method is smaller than that of the TOAmethod in the range of 0ndash12m With the increase ofmeasurement distance the localization error of the RSSImethod is larger than that of the TOA method when themeasurement distance is more than 12m -e RSSI-TOAmethod proposed in this paper has the smallest averagelocalization error under the same conditions

55 Localization Error of Unknown Nodes Figure 9 showsthe localization errors of 180 unknown nodes obtained by

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30Anchor node density ()

RSSI methodTOA methodRSSI-TOA method

Figure 5 Relationship between localization error and anchor nodedensity with mobile anchor nodes

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30Anchor node density ()

RSSI methodTOA methodRSSI-TOA method

Figure 6 Relationship between localization error and anchor nodedensity without mobile anchor node

151413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

RSSI methodTOA methodRSSI-TOA method

10 15 20 25 30 35 40Communication radius (m)

Figure 7 Relationship between communication radius and av-erage localization error

6 Journal of Electrical and Computer Engineering

the three methods It can be seen from Figure 9 that the nodelocalization errors of RSSI method and TOA method fluc-tuate greatly and there are many nodes having large errorsHowever the curve fluctuation of the node localization errorof the RSSI-TOAmethod proposed in this paper is relativelysmall and the localization error is the smallest among thethree methods It shows that in the three-dimensional en-vironment the RSSI-TOA method proposed in this paper is

the least affected by the external environment and has thehighest localization accuracy and the best localizationstability

6 Conclusion

In this work a ranging method is proposed that combinesRSSI and TOA ranging information in this paper -e

151617181920

1413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30 35 40Measuring position (m)

RSSI methodTOA methodRSSI-TOA method

Figure 8 Relationship between localization error and measurement position

RSSI methodTOA methodRSSI-TOA method

25

20

15

10

5

0

Aver

age l

ocal

izat

ion

erro

r (m

)

0 20 40 60 80 100 120 140 160 180Node number

Figure 9 Localization error of each node in three localization algorithms

Journal of Electrical and Computer Engineering 7

proposed ranging method reduces the ranging errors byusing RSSI or TOA ranging step by step in the ranging stageIn order to further improve the accuracy of node localiza-tion a mobile anchor node is introduced-e anchor node ismoved in the three-dimensional environment by GaussndashMarkov to form virtual anchor nodes -en the maximum-likelihood estimation method is used to realize the three-dimensional localization of WSN nodes -e simulationresults verified the effectiveness of the proposed RSSI-TOAmethod and achieved better localization results Because theRSSI ranging value and the TOA ranging value are greatlyaffected by WSN node devices and external environmentthe difference of the different node devices in differentapplication environments at different times will be relativelylarge Research on LSSVR node localization algorithm thatcombines RSSI and TOA ranging information in differentenvironments will be our future research At the same timethe static nodes are located in the three-dimensionalmonitoring area through amobile anchor node in this paper-e next step is to study the method of localization of themobile node by multiple mobile anchor nodes

Data Availability

Data sharing is not applicable to this article as no datasetswere generated or analysed during the current study

Conflicts of Interest

-e authors declare that they have no conflicts of interest

Acknowledgments

-is work was supported by the National Natural ScienceFoundation of China (no 61741303) Guangxi NaturalScience Foundation (no 2017GXNSFAA198161) KeyLaboratory of Spatial Information and Geomatics (GuilinUniversity of Technology) (nos 15-140-07-23 and 16-380-25-23) and Innovation Project of Guangxi Graduate Edu-cation (noYCSW2019159)

References

[1] R E Mohamed A I Saleh M Abdelrazzak and A S SamraldquoSurvey on wireless sensor network applications and energyefficient routing protocolsrdquo Wireless Personal Communica-tions vol 101 no 2 pp 1019ndash1055 2018

[2] K Langendoen and N Reijers ldquoDistributed localization inwireless sensor networksa quantitative comparisonrdquo Com-puter Networks vol 43 no 4 pp 499ndash518 2013

[3] G Han J Jiang C Zhang T Q Duong M Guizani andG K Karagiannidis ldquoA survey on mobile anchor nodeassisted localization in wireless sensor networksrdquo IEEECommunications Surveys amp Tutorials vol 18 no 3pp 2220ndash2243 2016

[4] K Han H Xing Z Deng and Y Du ldquoA RSSIPDR-basedprobabilistic position selection algorithm with NLOS iden-tification for indoor localisationrdquo ISPRS International Journalof Geo-Information vol 7 no 6 pp 232ndash253 2018

[5] C Angelo and A Fascista ldquoOn the hybrid TOARSS rangeestimation in wireless sensor networksrdquo IEEE Transactions onWireless Communications vol 17 no 1 pp 361ndash371 2018

[6] S Tomic M Beko M Tuba and V M F Correia ldquoTargetlocalization in NLOS environments using RSS and TOAmeasurementsrdquo IEEE Wireless Communications Lettersvol 7 no 6 pp 1062ndash1065 2018

[7] E Erdemir and T E Tuncer ldquoPath planning for mobile-anchor based wireless sensor network localization static anddynamic schemesrdquo Ad Hoc Networks vol 77 pp 1ndash10 2018

[8] Y Chen S Lu J Chen and T Ren ldquoNode localization al-gorithm of wireless sensor networks with mobile beaconnoderdquo Peer-to-Peer Networking and Applications vol 10no 3 pp 1ndash13 2016

[9] L Karim N Nasser Q H Mahmoud A Anpalagan andT E Salti ldquoRange-free localization approach for M2Mcommunication system using mobile anchor nodesrdquo Journalof Network and Computer Applications vol 47 pp 137ndash1462015

[10] Y Zhao J Xu and J Jiang ldquoRSSI based localization withmobile anchor for wireless sensor networksrdquo in Proceedingsof the 2017 International Conference on Geo-Spatial Knowledgeand Intelligence pp 176ndash187 Chiang Mai -ailand De-cember 2018

[11] P Singh A Khosla A Kumar and M Khosla ldquoOptimizedlocalization of target nodes using single mobile anchor node inwireless sensor networkrdquo AEUmdashInternational Journal ofElectronics and Communications vol 91 pp 55ndash65 2018

[12] P Singh A Khosla A Kumar and M Khosla ldquoComputa-tional intelligence based localization of moving target nodesusing single anchor node in wireless sensor networksrdquoTelecommunication Systems vol 69 no 3 pp 397ndash411 2018

[13] X C Song Y S Zhao and L ZWang ldquoGaussndashMarkov-basedmobile anchor localization (GM-MAL) algorithm based onlocal linear embedding optimization in internet of sensornetworksrdquo Cognitive Systems Research vol 52 pp 138ndash1432018

[14] T Das and S Roy ldquoEnergy efficient and event driven mobilitymodel in mobile WSNrdquo in Proceedings of the IEEE In-ternational Conference on Advanced Networks amp Telecom-munications Systems pp 1ndash6 Kolkata India December 2015

[15] Z Zhong D-Y Luo S-Q Liu X-P Fan and Z-H Qu ldquoAnadaptive localization approach for wireless sensor networksbased on gauss-markov mobility modelrdquo Acta AutomaticaSinica vol 36 no 11 pp 1557ndash1568 2010

[16] V Bianchi P Ciampolini and I De Munari ldquoRSSI-basedindoor localization and identification for zigbee wirelesssensor networks in smart homesrdquo IEEE Transactions on In-strumentation and Measurement vol 68 no 2 pp 566ndash5752019

[17] F Despaux K Jaffres-Runser A V D Bossche and T ValldquoAccurate and platform-agnostic time-of-flight estimation inultra-wide bandrdquo in Proceeding of the 27th IEEE InternationalSymposium on Personal Indoor and Mobile Radio Commu-nications pp 1ndash7 Valencia Spain September 2016

[18] S Sadowski and P Spachos ldquoRSSI-based indoor localizationwith the Internet of thingsrdquo IEEEAccess vol 6 pp 30149ndash301612018

[19] S Wu S Zhang and D Huang ldquoA TOA-based localizationalgorithm with simultaneous NLOS mitigation and syn-chronization error eliminationrdquo IEEE Sensors Letters vol 3no 3 pp 1ndash4 2019

8 Journal of Electrical and Computer Engineering

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

the three methods It can be seen from Figure 9 that the nodelocalization errors of RSSI method and TOA method fluc-tuate greatly and there are many nodes having large errorsHowever the curve fluctuation of the node localization errorof the RSSI-TOAmethod proposed in this paper is relativelysmall and the localization error is the smallest among thethree methods It shows that in the three-dimensional en-vironment the RSSI-TOA method proposed in this paper is

the least affected by the external environment and has thehighest localization accuracy and the best localizationstability

6 Conclusion

In this work a ranging method is proposed that combinesRSSI and TOA ranging information in this paper -e

151617181920

1413121110

9876543210

Aver

age l

ocal

izat

ion

erro

r (m

)

5 10 15 20 25 30 35 40Measuring position (m)

RSSI methodTOA methodRSSI-TOA method

Figure 8 Relationship between localization error and measurement position

RSSI methodTOA methodRSSI-TOA method

25

20

15

10

5

0

Aver

age l

ocal

izat

ion

erro

r (m

)

0 20 40 60 80 100 120 140 160 180Node number

Figure 9 Localization error of each node in three localization algorithms

Journal of Electrical and Computer Engineering 7

proposed ranging method reduces the ranging errors byusing RSSI or TOA ranging step by step in the ranging stageIn order to further improve the accuracy of node localiza-tion a mobile anchor node is introduced-e anchor node ismoved in the three-dimensional environment by GaussndashMarkov to form virtual anchor nodes -en the maximum-likelihood estimation method is used to realize the three-dimensional localization of WSN nodes -e simulationresults verified the effectiveness of the proposed RSSI-TOAmethod and achieved better localization results Because theRSSI ranging value and the TOA ranging value are greatlyaffected by WSN node devices and external environmentthe difference of the different node devices in differentapplication environments at different times will be relativelylarge Research on LSSVR node localization algorithm thatcombines RSSI and TOA ranging information in differentenvironments will be our future research At the same timethe static nodes are located in the three-dimensionalmonitoring area through amobile anchor node in this paper-e next step is to study the method of localization of themobile node by multiple mobile anchor nodes

Data Availability

Data sharing is not applicable to this article as no datasetswere generated or analysed during the current study

Conflicts of Interest

-e authors declare that they have no conflicts of interest

Acknowledgments

-is work was supported by the National Natural ScienceFoundation of China (no 61741303) Guangxi NaturalScience Foundation (no 2017GXNSFAA198161) KeyLaboratory of Spatial Information and Geomatics (GuilinUniversity of Technology) (nos 15-140-07-23 and 16-380-25-23) and Innovation Project of Guangxi Graduate Edu-cation (noYCSW2019159)

References

[1] R E Mohamed A I Saleh M Abdelrazzak and A S SamraldquoSurvey on wireless sensor network applications and energyefficient routing protocolsrdquo Wireless Personal Communica-tions vol 101 no 2 pp 1019ndash1055 2018

[2] K Langendoen and N Reijers ldquoDistributed localization inwireless sensor networksa quantitative comparisonrdquo Com-puter Networks vol 43 no 4 pp 499ndash518 2013

[3] G Han J Jiang C Zhang T Q Duong M Guizani andG K Karagiannidis ldquoA survey on mobile anchor nodeassisted localization in wireless sensor networksrdquo IEEECommunications Surveys amp Tutorials vol 18 no 3pp 2220ndash2243 2016

[4] K Han H Xing Z Deng and Y Du ldquoA RSSIPDR-basedprobabilistic position selection algorithm with NLOS iden-tification for indoor localisationrdquo ISPRS International Journalof Geo-Information vol 7 no 6 pp 232ndash253 2018

[5] C Angelo and A Fascista ldquoOn the hybrid TOARSS rangeestimation in wireless sensor networksrdquo IEEE Transactions onWireless Communications vol 17 no 1 pp 361ndash371 2018

[6] S Tomic M Beko M Tuba and V M F Correia ldquoTargetlocalization in NLOS environments using RSS and TOAmeasurementsrdquo IEEE Wireless Communications Lettersvol 7 no 6 pp 1062ndash1065 2018

[7] E Erdemir and T E Tuncer ldquoPath planning for mobile-anchor based wireless sensor network localization static anddynamic schemesrdquo Ad Hoc Networks vol 77 pp 1ndash10 2018

[8] Y Chen S Lu J Chen and T Ren ldquoNode localization al-gorithm of wireless sensor networks with mobile beaconnoderdquo Peer-to-Peer Networking and Applications vol 10no 3 pp 1ndash13 2016

[9] L Karim N Nasser Q H Mahmoud A Anpalagan andT E Salti ldquoRange-free localization approach for M2Mcommunication system using mobile anchor nodesrdquo Journalof Network and Computer Applications vol 47 pp 137ndash1462015

[10] Y Zhao J Xu and J Jiang ldquoRSSI based localization withmobile anchor for wireless sensor networksrdquo in Proceedingsof the 2017 International Conference on Geo-Spatial Knowledgeand Intelligence pp 176ndash187 Chiang Mai -ailand De-cember 2018

[11] P Singh A Khosla A Kumar and M Khosla ldquoOptimizedlocalization of target nodes using single mobile anchor node inwireless sensor networkrdquo AEUmdashInternational Journal ofElectronics and Communications vol 91 pp 55ndash65 2018

[12] P Singh A Khosla A Kumar and M Khosla ldquoComputa-tional intelligence based localization of moving target nodesusing single anchor node in wireless sensor networksrdquoTelecommunication Systems vol 69 no 3 pp 397ndash411 2018

[13] X C Song Y S Zhao and L ZWang ldquoGaussndashMarkov-basedmobile anchor localization (GM-MAL) algorithm based onlocal linear embedding optimization in internet of sensornetworksrdquo Cognitive Systems Research vol 52 pp 138ndash1432018

[14] T Das and S Roy ldquoEnergy efficient and event driven mobilitymodel in mobile WSNrdquo in Proceedings of the IEEE In-ternational Conference on Advanced Networks amp Telecom-munications Systems pp 1ndash6 Kolkata India December 2015

[15] Z Zhong D-Y Luo S-Q Liu X-P Fan and Z-H Qu ldquoAnadaptive localization approach for wireless sensor networksbased on gauss-markov mobility modelrdquo Acta AutomaticaSinica vol 36 no 11 pp 1557ndash1568 2010

[16] V Bianchi P Ciampolini and I De Munari ldquoRSSI-basedindoor localization and identification for zigbee wirelesssensor networks in smart homesrdquo IEEE Transactions on In-strumentation and Measurement vol 68 no 2 pp 566ndash5752019

[17] F Despaux K Jaffres-Runser A V D Bossche and T ValldquoAccurate and platform-agnostic time-of-flight estimation inultra-wide bandrdquo in Proceeding of the 27th IEEE InternationalSymposium on Personal Indoor and Mobile Radio Commu-nications pp 1ndash7 Valencia Spain September 2016

[18] S Sadowski and P Spachos ldquoRSSI-based indoor localizationwith the Internet of thingsrdquo IEEEAccess vol 6 pp 30149ndash301612018

[19] S Wu S Zhang and D Huang ldquoA TOA-based localizationalgorithm with simultaneous NLOS mitigation and syn-chronization error eliminationrdquo IEEE Sensors Letters vol 3no 3 pp 1ndash4 2019

8 Journal of Electrical and Computer Engineering

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

proposed ranging method reduces the ranging errors byusing RSSI or TOA ranging step by step in the ranging stageIn order to further improve the accuracy of node localiza-tion a mobile anchor node is introduced-e anchor node ismoved in the three-dimensional environment by GaussndashMarkov to form virtual anchor nodes -en the maximum-likelihood estimation method is used to realize the three-dimensional localization of WSN nodes -e simulationresults verified the effectiveness of the proposed RSSI-TOAmethod and achieved better localization results Because theRSSI ranging value and the TOA ranging value are greatlyaffected by WSN node devices and external environmentthe difference of the different node devices in differentapplication environments at different times will be relativelylarge Research on LSSVR node localization algorithm thatcombines RSSI and TOA ranging information in differentenvironments will be our future research At the same timethe static nodes are located in the three-dimensionalmonitoring area through amobile anchor node in this paper-e next step is to study the method of localization of themobile node by multiple mobile anchor nodes

Data Availability

Data sharing is not applicable to this article as no datasetswere generated or analysed during the current study

Conflicts of Interest

-e authors declare that they have no conflicts of interest

Acknowledgments

-is work was supported by the National Natural ScienceFoundation of China (no 61741303) Guangxi NaturalScience Foundation (no 2017GXNSFAA198161) KeyLaboratory of Spatial Information and Geomatics (GuilinUniversity of Technology) (nos 15-140-07-23 and 16-380-25-23) and Innovation Project of Guangxi Graduate Edu-cation (noYCSW2019159)

References

[1] R E Mohamed A I Saleh M Abdelrazzak and A S SamraldquoSurvey on wireless sensor network applications and energyefficient routing protocolsrdquo Wireless Personal Communica-tions vol 101 no 2 pp 1019ndash1055 2018

[2] K Langendoen and N Reijers ldquoDistributed localization inwireless sensor networksa quantitative comparisonrdquo Com-puter Networks vol 43 no 4 pp 499ndash518 2013

[3] G Han J Jiang C Zhang T Q Duong M Guizani andG K Karagiannidis ldquoA survey on mobile anchor nodeassisted localization in wireless sensor networksrdquo IEEECommunications Surveys amp Tutorials vol 18 no 3pp 2220ndash2243 2016

[4] K Han H Xing Z Deng and Y Du ldquoA RSSIPDR-basedprobabilistic position selection algorithm with NLOS iden-tification for indoor localisationrdquo ISPRS International Journalof Geo-Information vol 7 no 6 pp 232ndash253 2018

[5] C Angelo and A Fascista ldquoOn the hybrid TOARSS rangeestimation in wireless sensor networksrdquo IEEE Transactions onWireless Communications vol 17 no 1 pp 361ndash371 2018

[6] S Tomic M Beko M Tuba and V M F Correia ldquoTargetlocalization in NLOS environments using RSS and TOAmeasurementsrdquo IEEE Wireless Communications Lettersvol 7 no 6 pp 1062ndash1065 2018

[7] E Erdemir and T E Tuncer ldquoPath planning for mobile-anchor based wireless sensor network localization static anddynamic schemesrdquo Ad Hoc Networks vol 77 pp 1ndash10 2018

[8] Y Chen S Lu J Chen and T Ren ldquoNode localization al-gorithm of wireless sensor networks with mobile beaconnoderdquo Peer-to-Peer Networking and Applications vol 10no 3 pp 1ndash13 2016

[9] L Karim N Nasser Q H Mahmoud A Anpalagan andT E Salti ldquoRange-free localization approach for M2Mcommunication system using mobile anchor nodesrdquo Journalof Network and Computer Applications vol 47 pp 137ndash1462015

[10] Y Zhao J Xu and J Jiang ldquoRSSI based localization withmobile anchor for wireless sensor networksrdquo in Proceedingsof the 2017 International Conference on Geo-Spatial Knowledgeand Intelligence pp 176ndash187 Chiang Mai -ailand De-cember 2018

[11] P Singh A Khosla A Kumar and M Khosla ldquoOptimizedlocalization of target nodes using single mobile anchor node inwireless sensor networkrdquo AEUmdashInternational Journal ofElectronics and Communications vol 91 pp 55ndash65 2018

[12] P Singh A Khosla A Kumar and M Khosla ldquoComputa-tional intelligence based localization of moving target nodesusing single anchor node in wireless sensor networksrdquoTelecommunication Systems vol 69 no 3 pp 397ndash411 2018

[13] X C Song Y S Zhao and L ZWang ldquoGaussndashMarkov-basedmobile anchor localization (GM-MAL) algorithm based onlocal linear embedding optimization in internet of sensornetworksrdquo Cognitive Systems Research vol 52 pp 138ndash1432018

[14] T Das and S Roy ldquoEnergy efficient and event driven mobilitymodel in mobile WSNrdquo in Proceedings of the IEEE In-ternational Conference on Advanced Networks amp Telecom-munications Systems pp 1ndash6 Kolkata India December 2015

[15] Z Zhong D-Y Luo S-Q Liu X-P Fan and Z-H Qu ldquoAnadaptive localization approach for wireless sensor networksbased on gauss-markov mobility modelrdquo Acta AutomaticaSinica vol 36 no 11 pp 1557ndash1568 2010

[16] V Bianchi P Ciampolini and I De Munari ldquoRSSI-basedindoor localization and identification for zigbee wirelesssensor networks in smart homesrdquo IEEE Transactions on In-strumentation and Measurement vol 68 no 2 pp 566ndash5752019

[17] F Despaux K Jaffres-Runser A V D Bossche and T ValldquoAccurate and platform-agnostic time-of-flight estimation inultra-wide bandrdquo in Proceeding of the 27th IEEE InternationalSymposium on Personal Indoor and Mobile Radio Commu-nications pp 1ndash7 Valencia Spain September 2016

[18] S Sadowski and P Spachos ldquoRSSI-based indoor localizationwith the Internet of thingsrdquo IEEEAccess vol 6 pp 30149ndash301612018

[19] S Wu S Zhang and D Huang ldquoA TOA-based localizationalgorithm with simultaneous NLOS mitigation and syn-chronization error eliminationrdquo IEEE Sensors Letters vol 3no 3 pp 1ndash4 2019

8 Journal of Electrical and Computer Engineering

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom