An Efficient SC-FDM Modulation Technique for a UAV Communication Link

electronics Article An Efficient SC-FDM Modultion Technique for UAV Communiction Link Sukhrob Aev 1, Oh-Heum Kwon 1, Suk-Hwn Lee 2 Ki-Ryong Kwon 1, * 1 Deprtment of IT Convergence Appliction Engineering, Pukyong Ntionl University, Busn 48513, Kore; sukhrobreus@pukyong.c.kr (S.A.); ohheum@gmil.com (O.-H.K.) 2 Deprtment of Informtion Security, Tongmyong University, Busn 48520, Kore; skylee@tu.c.kr * Correspondence: krkwon@pknu.c.kr; Tel.: +82-51-629-6257 Received: 25 September 2018; Accepted: 21 November 2018; Published: 25 November 2018 Abstrct: Since communiction link of n unmnned eril vehicle (UAV) its relibility evlution represent n rduous field, we hve concentrted our work on this pic. The dem regrding vlidity relibility of communiction dt link of UAV is much higher since environment of modern bttlefield is becoming more more complex. Therefore, communiction chnnel between vehicle ground control sttion (GCS) should be secure provide n efficient dt link. In ddition, similr or types of communictions, dt link of UAV hs severl requirements such s long-rnge opertion, high efficiency, relibility, low ltency. In order chieve n efficient dt link, we need dopt highly efficient modultion technique, which leds n increse in flight time of UAV, dt trnsmission rte, relibility of communiction link. For this purpose, we hve investigted single-crrier division multiplexing (SC-FDM) modultion technique for UAV communiction system. The results obtined from comprtive study demonstrte tht SC-FDM hs better performnce thn currently used modultion technique for UAV communiction link. We expect tht our proposed pproch cn be remrkble frmework tht will help drone mnufcturers estblish n efficient UAV communiction link extend flight durtion of drones, especilly those being used for serch rescue opertions, militry tsks, delivery services. Keywords: UAV communiction system; dt link; SC-FDM; pek--verge power rtio (PAPR); modultion 1. Introduction With rpidly dvncing technology, unmnned eril vehicles (UAVs), widely known s drones, re becoming incresingly effective significntly less costly during recent yers. These vehicles cn be controlled eir under remote control (RC) by pilot operr or unomously by onbord computers. A UAV communiction chnnel is key fcr tht cn ffect performnce of dt link in terms of high dt rte relible trnsmission of informtion. In or words, ensuring efficiency of UAV communiction link represents one of gret chllenges of current works regrding UAV communiction system. The UAV communiction system hs following mjor requirements: Efficient dt link Long-rnge opertion Bidirectionl communiction Low ltency Long flight time Electronics 2018, 7, 352; doi:10.3390/electronics7120352 www.mdpi.com/journl/electronics

Electronics 2018, 7, 352 2 of 18 Opertionl cpbilities Relible communiction As trnsmission of control comms gred dt, which cn be recorded video phos, is chieved through communiction chnnel between UAV ground control sttion (GCS), UAV dt link requires highest relibility in dt trnsmission s well s high dt trnsfer rte. Nowdys, in order provide n efficient communiction link, mny drones use spred spectrum technology tht llows mny different pilots operte in sme 2.4 GHz b without conflicts. Receivers in this b re virtully immune ginst interference issues. Essentilly, two types of spred spectrum technology re used for UAV communiction link. The first one is -hopping spred spectrum (FHSS), which uncesingly chnges its nrrowb on severl occsions second within 2.4 GHz rnge. In this process, receiver recognizes ptterns of tht re utilized by trnsmitter. Becuse trnsmitter chnges from one nor, receiver cn dopt suitble. Unlike FHSS, direct-sequence spred spectrum (DSSS) system uses much wider bwidth trnsmit signl on single selected. The trnsmitter sends n originl nrrowb signl vi spreding code generr tht multiplies nrrowb dt signl using much higher. Anywy, both spred spectrum modes (FHSS DSSS) trnsmit signl within 2.4 GHz b. In prctice, orthogonl division multiplexing (OFDM) modultion hs been considered more efficient thn FHSS DSSS due its greter lernce of multipth disrtion, higher throughput, potentil dt rte [1]. Nowdys, dem on long-rnge opertion long flight time is incresing in UAV communiction systems currently used modultion techniques hve fundmentl constrint meet this dem. Therefore, we need dopt potentil modultion technique tht fulfills this dem. In fct, it is essentil dopt most effective technique tht cn provide highly efficient dt link between vehicle GCS. The min contribution of this pper is investigtion of SC-FDM modultion technique in UAV communiction system in order provide n efficient UAV communiction link extend flight time (bttery life) of drones. The performnce of dopted modultion technique is nlyzed by compring it with OFDM modultion. Additionlly, SC-FDM hs been considered trnsfer dt using different kinds of modultion schemes such s M-ry phse-shift keying (BPSK, QPSK, 8-PSK) M-ry qudrture mplitude modultion (16-QAM 64-QAM) in this work. The remining prt of our pper is structured s follows. In Section 2, some relted works re discussed. Section 3 demonstrtes UAV communiction link communiction system components. The comprison between proposed modultion technique OFDM modultion is presented in Section 4. We show experimentl setup performnce mesures of system in Section 5. Afterwrds, Section 6 illustrtes comprtive results obtined from experiments. At end of pper, Section 7 presents our conclusions future work. 2. Relted Works In recent yers, extending bttery life flight time of qudcopters hs become crucil tsk, since most of se vehicles re used for delivery services militry tsks. To mintin flight time, qudcopter power modeling is bsic technology becuse limittion of flight time ctully comes from bttery cpcity constrint. Mekw et l. [2] proposed simple model of power consumption for delivery qudcopters by testing Prrot AR. Drone 2.0 on horizontl flight. The power consumption of drone ws mesured by current logger light weighted voltge. However, ir proposed power model is bsed only on verge power consumption dt obtined during horizontl flight. The work by Sowh et l. [3] presents rottionl energy hrvester powered by rors using brushless dc (BLDC) generr increse flight time of qudcopters. A printed circuit bord (PCB) Egle PCB design softwre (EAGLE 6.4.0 Light) were used build physicl model. The hrvester interfce circuit consumed 1.5 3.2 V s n input from rectifier

Electronics 2018, 7, 352 3 of 18 circuit produced n djustble output of 18 V with 2.7 W output power for ech generr t Electronics 2018, 7, x FOR PEER REVIEW 3 of 18 82% efficiency. A 600 ma of current from four generrs ws utilized provide extr flight time extr for flight time qudcopter, for thus, qudcopter, giningthus, boutgining 42% in bout flight durtion. 42% in flight Moreover, durtion. in [4], Moreover, comprison in [4], of BLDC comprison control of BLDC field-oriented control control field-oriented (FOC) techniques control (FOC) hs techniques been nlyzed hs in been order nlyzed enhnce in order flight enhnce endurnce flight of multiror endurnce UAVs. of multiror According UAVs. According power efficiency power output efficiency rque qulity output of rque propeller qulity electricl of propeller drives, electricl FOC hs drives, shown FOC better hs performnce shown better by performnce 2 4% in efficiency by 2 4% compred in efficiency BLDC compred control, BLDC leding control, flight leding endurnce flight improvement. endurnce improvement. A method for forimproving efficiency of of UAV UAVcommuniction link link cn cn be be found found in [5]. in [5]. In In this this work, work, key keytsks tsks of of UAV UAVcommuniction system system chrcteristics of of rdio chnnel between UAV ground control unit (GCU) hve been nlyzed. Considering vrious issues ssocited with with UAV UAV communiction communiction link, n link, uthor n proposed uthor proposed optiml rdio optiml chnnel rdio construction chnnel using construction rotry using mounting rotry pltform mounting with pltform ntenns, with power ntenns, power low-noise mplifiers low-noise mplifiers s well s n s well OFDM s n modultion OFDM modultion technique technique increse increse dt trnsfer dt rte. trnsfer In [6], rte. Wu In et [6], l. Wu proposed et l. proposed OFDM s OFDM trnsmission s trnsmission system system for UAVfor wireless UAV wireless communictions. communictions. Initilly, Initilly, find out find proper out OFDM proper system OFDM prmeters, system prmeters, coherence coherence time time Doppler spred Doppler hve spred beenhve mesured. been mesured. After obtining After obtining inter-crrier inter-crrier interference interference (ICI) coefficients, (ICI) coefficients, y evluted y evluted bit error rte bit (BER) error performnce rte (BER) of performnce OFDM technology of OFDM intechnology typicl UAV in communiction typicl UAV communiction chnnel se chnnel performnce se performnce results were compred results were with compred those ofwith OFDM those in of norml OFDM wireless in norml indoor wireless chnnels. indoor According chnnels. According ir simultion ir results, simultion n insignificnt results, insignificnt performnce performnce degrdtion degrdtion cn be seencn when be seen OFDM when technology OFDM technology is pplied is pplied UAV communiction UAV communiction chnnel. chnnel. In UAV pplictions, re re hve hve been been few few efforts efforts dopt dopt SC-FDM SC-FDM modultion modultion technology technology for for UAV UAV communiction communiction system, system, while while it hs it hs been been widely widely ccepted ccepted in mobile in mobile communictions. communictions. In [7], In Miko [7], Miko Nemeth Nemeth proposed proposed hrdwre hrdwre rchitecture rchitecture which which includes includes Xilinx Xilinx field-progrmmble field-progrmmble gte rry gte rry (FPGA) (FPGA) combined combined with with softwre-defined softwre-defined rdio (SDR) rdio chip (SDR) chip SC-FDM SC-FDM modultion modultion system system provide provide high dt high trnsmission dt trnsmission rte rte rdio nvigtion rdio nvigtion for for communiction communiction link of link UAV of systems. UAV systems. Their proposed Their proposed hrdwre hrdwre design ofdesign trnsceiver of trnsceiver is shown inis Figure shown 1. They in Figure implemented 1. They modultion, implemented demodultion, demodultion, coding functions coding functions FPGA. However, in FPGA. re is However, no indiction re of is no performnce indiction of of SC-FDM performnce modultion of SC-FDM in ir modultion work. in ir work. Figure 1. Hrdwre rchitecture. Figure 1. Hrdwre rchitecture. Until now, most of previous works on single-crrier division multiple Until now, most of previous works on single-crrier division multiple ccess ccess (SC-FDMA) hve been crried out for uplink communictions in long-term evolution (SC-FDMA) hve been crried out for uplink communictions in long-term evolution (LTE) (LTE) technology of mobile communiction systems [8 13]. As n lterntive orthogonl technology of mobile communiction systems [8 13]. As n lterntive orthogonl division multiple ccess (OFDMA), SC-FDMA hs drwn considerble ttention in mobile division multiple ccess (OFDMA), SC-FDMA hs drwn considerble ttention in mobile communictions. In [14], Myung gives n overview of SC-FDMA. Anor reserch focuses on PAPR communictions. In [14], Myung gives n overview of SC-FDMA. Anor reserch focuses on PAPR reduction of loclized SC-FDMA using prtil trnsmit sequence (PTS) [15]. Actully, re cn be reduction of loclized SC-FDMA using prtil trnsmit sequence (PTS) [15]. Actully, re cn be loclized distributed modes of subcrrier mpping in SC-FDMA [14,16]. In loclized SC-FDMA, loclized distributed modes of subcrrier mpping in SC-FDMA [14,16]. In loclized SC-FDMA, ech terminl uses set of contiguous subcrriers for trnsmission of symbols, reby limiting ech terminl uses set of contiguous subcrriers for trnsmission of symbols, reby limiting m only portion of system bwidth. On or h, subcrriers used by terminl m only portion of system bwidth. On or h, subcrriers used by re propgted throughout entire bwidth in distributed SC-FDMA. Loclized SC-FDMA, terminl re propgted throughout entire bwidth in distributed SC-FDMA. Loclized which is used for uplink trnsmission of LTE systems, hs lower PAPR thn distributed SC-FDMA. SC-FDMA, which is used for uplink trnsmission of LTE systems, hs lower PAPR thn distributed Furrmore, Tsiropoulou et l. [17] provided brgining model power optimiztion SC-FDMA. frmework solve problem of subcrrier power lloction in multiuser SC-FDMA wireless Furrmore, Tsiropoulou et l. [17] provided brgining model power optimiztion networks. The obtined numericl results key fetures of ir proposed pproch demonstrte frmework solve problem of subcrrier power lloction in multiuser SC-FDMA wireless tht introduced frmework cn be foundtion for supporting heterogeneous services networks. The obtined numericl results key fetures of ir proposed pproch demonstrte tht introduced frmework cn be foundtion for supporting heterogeneous services implementtion of different users priorities ccess vilble resources. Towrds this direction, similr work cn be found in [18]. In 2016, Tsiropoulou et l. [19] studied exmined

Electronics 2018, 7, 352 4 of 18 implementtion Electronics 2018, 7, x FOR of different PEER REVIEW users priorities ccess vilble resources. Towrds this direction, 4 of 18 similr work cn be found in [18]. In 2016, Tsiropoulou et l. [19] studied exmined vrious stte-of--rt vrious stte-of--rt resource lloction resource lgorithms lloction lgorithms frmeworks frmeworks developed developed llocte llocte subcrriers subcrriers trnsmission trnsmission power of users power in of users uplink of SC-FDMA uplink of wireless SC-FDMA networks. wireless Luo networks. Xiong Luo [20] proposed Xiong [20] proposed SC-FDMA-IDMA SC-FDMA-IDMA system model, system which model, is which combintion is combintion of SC-FDMAof SC-FDMA interleved division interleved multiple division ccess multiple (IDMA) ccess (IDMA) studied studied effect of crrier effect of crrier offset (CFO) offset on (CFO) BER performnce on BER performnce of this system of model. this system model. 3. UAV Communiction System In generl, UAV communiction link cn both send control comms from GCS vehicle receive dt bout flight flight on on downlink, s s shown shown in in Figure Figure 2. 2. A A bidirectionl bidirectionl link link cn cn be estblished be estblished in order in order provide provide communiction communiction between between drone GCS drone [21]. A communiction GCS [21]. A link communiction between se link two between components se two hs components provide long-rnge hs provide opertions long-rnge s wellopertions s continuous s well s stble continuous link. Therefore, stble link. estblishment Therefore, ofestblishment chnnel model of tht chnnel is suitble model for tht UAV is suitble chrcteristics for UAV plys chrcteristics n importnt plys role n importnt in improving role in improving dt link of dt UAV link [22,23]. of Furrmore, UAV [22,23]. for Furrmore, improving relibility for improving of relibility dt link, of n dptive dt link, informtion n dptive rte informtion method is presented rte method in is [24]. presented in [24]. Figure 2. Drone communiction link. Figure Figure 3 illustrtes illustrtes components components of of UAV UAV communiction communiction system. system. The The min min component component is is microcontroller, microcontroller, lso lso referred referred s s flight flight controller, controller, which which is is core core for ll for functioning ll functioning of UAV. of It UAV. mnges It mnges filsfe, upilot, filsfe, upilot, wypoints, wypoints, mny or unomous mny or functions. unomous This functions. microcontroller This interprets microcontroller input interprets from receiver, input from globl positioning receiver, globl system positioning (GPS) module, system bttery (GPS) monir, module, inertil bttery mesurement monir, inertil unit mesurement (IMU) tht unit include (IMU) n tht ccelerometer include n ccelerometer gyroscope which gyroscope cn be which used cn for providing be used for stbility providing or stbility mintin or mintin reference direction reference in direction nvigtion in nvigtion system [25], system or [25], onbord or sensors. onbord The sensors. GCS The provides GCS provides relevnt dt relevnt bout dt bout vehicle such vehicle s speed, such s ttitude, speed, ttitude, ltitude, loction, ltitude, yw, loction, pitch, yw, roll, pitch, wrnings, roll, wrnings, or informtion or informtion [26]. As [26]. shown As shown in Figure in Figure 3, UAV 3, hs UAV two hs links: two dt links: link dt link communiction communiction link. To link. ensure To ensure trnsmission trnsmission of dt of between dt between vehicle vehicle GCS, GCS, UAV uses UAV uses dt link dt tht link opertes tht opertes in in rnge rnge from 150 from MHz 150 MHz 1.5 GHz. 1.5 On GHz. On or h, or h, 2.4 GHz 2.4 GHz b tht determines b tht determines communiction communiction link between link trnsmitter between receiver trnsmitter is used receiver in order is used control in order vehicle. control It should vehicle. be noted It should tht be noted trnsmitter tht trnsmitter receiver must receiver both be must on both sme be. on sme In point. of fct, In drones point of hve fct, exclusive drones hve use of exclusive ir own use of ir lloction own due lloction longer due rnge potentilly longer rnge worse consequences potentilly worse of rdio consequences interference. of Initilly, rdio drones interference. scn Initilly, rnge of drones frequencies scn within rnge of frequencies 2.4 GHz b within use only 2.4 GHz nrrowb b use only tht nrrowb is not in use by nor tht is drone. not in As use by result nor of this, drone. mny As drones result cn of this, utilize mny 2.4 drones GHz cn utilize b 2.4 GHz simultneously. b This simultneously. feture of drones This cn feture be noticeble of drones when cn be number noticeble of drones when re number used s flying of drones bse re sttions used s in flying wireless bse cellulr sttions networks in wireless serve cellulr n networks rbitrrily locted serve n set rbitrrily of users [27 29]. locted Moreover, set of users typicl [27 29]. UAVs Moreover, use multiple typicl UAVs rdio use interfces multiple rdio mintin interfces continuous mintin connection continuous with essentil connection links with essentil GCSs, or links UAVs, GCSs, or stellite UAVs, relys. stellite relys.

Electronics 2018, 7, x FOR PEER REVIEW 5 of 18 Electronics 2018, 7, 352 5 of 18 Electronics 2018, 7, x FOR PEER REVIEW 5 of 18 Figure 3. Unmnned eril vehicle (UAV) communiction system components. Figure Figure 3. 3. Unmnned Unmnned eril eril vehicle vehicle (UAV) (UAV) communiction communiction system system components. components. The modultion demodultion process of trnsmitted signl through wireless chnnel cn respectively The modultion be done in demodultion trnsmitter process receiver of trnsmitted of UAV signl communiction through wireless system. Figure chnnel 4 The modultion demodultion process of trnsmitted signl through wireless chnnel cn shows respectively typicl betrnsmitter done in of trnsmitter UAV communiction receiver of system UAV tht communiction trnsmits control system. comms Figure 4 cn respectively be done in trnsmitter receiver of UAV communiction system. Figure shows telemetry typicl dt. trnsmitter Initilly, of UAV input communiction dt is sred in system dt tht srge trnsmits module, control n, comms chnnel shows typicl trnsmitter of UAV communiction system tht trnsmits control comms coding telemetry cn be dt. used Initilly, for error correction input dt encoding. is sred After in dt tht, srge dt module, strems n, re mpped chnnelin coding telemetry dt. Initilly, input dt is sred in dt srge module, n, chnnel cn frmes be used redy for error for correction chnnel modultion. encoding. After The tht, bseb dt modultion strems re of mpped ech crrier in cn frmes be selected coding cn be used for error correction encoding. After tht, dt strems re mpped in redy mong forbpsk, chnnel QPSK, modultion. 8-PSK, The 16-QAM, bseb modultion 64-QAM depending of ech crrier on cn be chnnel selectedcondition. mong BPSK, The frmes redy for chnnel modultion. The bseb modultion of ech crrier cn be selected QPSK, modulted 8-PSK, signls 16-QAM, re n 64-QAM directly depending converted in chnnel rdio condition. The modulted b for signls wireless re mong BPSK, QPSK, 8-PSK, 16-QAM, 64-QAM depending on chnnel condition. The n trnsmission. directly converted Before trnsmitting in rdio signl, rdio b for wireless (RF) trnsmission. mplifier cn Before be used trnsmitting convert modulted signls re n directly converted in rdio b for wireless signl, low-power rdio signl (RF) in mplifier higher cnone. be used Finlly, convert RF signl low-power trnsmission is signl done t in trnsmission. Before trnsmitting signl, rdio (RF) mplifier cn be used convert higher trnsmitter one. ntenn. Finlly, RF signl trnsmission is done t trnsmitter ntenn. low-power signl in higher one. Finlly, RF signl trnsmission is done t trnsmitter ntenn. Figure 4. Block digrm of UAV trnsmitter. 4. Figure 4. Block digrm of UAV trnsmitter. The Proposed SC-FDM OFDM 4. The In Proposed generl, SC-FDM behves OFDM like single-crrier system with short symbol durtion compred OFDM. To chieve this, SC-FDM introduces n N-point discrete Fourier trnsform (DFT) block In generl, SC-FDM behves like single-crrier system with short symbol durtion compred right fter seril prllel converter in OFDM structure. The DFT block converts prllel OFDM. To chieve this, SC-FDM introduces n N-point discrete Fourier trnsform (DFT) block sequences of symbols in in time domin different points. The The mjor mjor disdvntge of right fter seril prllel converter in OFDM structure. The DFT block converts prllel of OFDM OFDM is its ishigh its high PAPR. PAPR. This This consequence is consequence of of fct tht fct tht trnsferred trnsferred signl is signl mount is sequences of symbols in time domin different points. The mjor disdvntge of of ll modulted subcrriers some of m re in phse with high mplitudes tht cnnot OFDM is its high PAPR. This is consequence of fct tht trnsferred signl is mount of ll modulted subcrriers some of m re in phse with high mplitudes tht cnnot

Electronics 2018, 7, 352 6 of 18 Electronics 2018, 7, x FOR PEER REVIEW 6 of 18 mount of ll modulted subcrriers some of m re in phse with high mplitudes tht cnnot be voided be voided [30]. Due [30]. Due this this reson, reson, power power structure structure of of trnsmitter is is chrcterized by by reltively low low verge- high-power peks. SC-FDM llows symbol be be trnsmitted in in prts over over multiple multiple subcrriers, subcrriers, but but in OFDM, in OFDM, we hve we one hve one mpping one mpping between symbol between symbol subcrrier. For subcrrier. exmple, For inexmple, OFDM one in OFDM symbolone occupies symbol one occupies subcrrier one subcrrier of 15 khz, of but 15 inkhz, SC-FDM, but in SC-FDM, sme symbol sme is symbol distributed distributed mong multiple mong subcrriers multiple subcrriers of 15 khz, of s15 described khz, s described in Figure 5 in [31]. Figure 5 [31]. Figure 5. Orthogonl division multiplexing (OFDM) single-crrier division Figure 5. Orthogonl division multiplexing (OFDM) single-crrier division multiplexing (SC-FDM). multiplexing (SC-FDM). 4.1. OFDM 4.1. OFDM In Figure 5, shown bove, dt cn be trnsferred by prllel subcrriers of 15 khz in OFDM. On In time Figure xis, 5, shown furr bove, divided dt subcrriers cn be trnsferred represent by blocks prllel of one subcrriers symbol durtion. of 15 khz This in OFDM. bsic unit On is known time xis, s resource furr element, divided subcrriers one symbol represent is crried blocks by of one one resource symbol element. durtion. InThis ddition, bsic unit number is known of resource s resource elements element, re used one mke symbol resource is crried block by tht one is resource bsic element. unit of scheduling. In ddition, At number beginning of resource of elements modultion re process, used mke dt resource is modulted block by tht is prticulr bsic modultion unit of scheduling. scheme in At order beginning trnsfer of dt modultion over se process, resource dt elements. is modulted This modultion by prticulr scheme modultion depends scheme on physicl in order chnnels trnsfer mpped dt on over se resource resource grid. elements. Then, M-point This modultion inverse discrete scheme Fourier depends trnsform on (IDFT) physicl trnsforms chnnels mpped signls on of resource prllel grid. Then, M-point domin in inverse smples discrete of Fourier composite trnsform time domin (IDFT) trnsforms signl, which re signls much of esier prllel generte t trnsmitter domin in side. smples All we need of composite do is send time se domin time signl, domin which smples re much t rdio esier frequencies. generte In t wireless trnsmitter chnnels, side. due All we multipth need propgtion, do is send re se cn time be domin dely spred smples t intersymbol rdio frequencies. interference In wireless (ISI) [32]. chnnels, This interference due multipth my cuse propgtion, given trnsmitted re cn be symbol dely spred be disrted intersymbol by or interference trnsmitted (ISI) symbols. [32]. This Since interference OFDM uses my composite cuse given IDFT smples, trnsmitted cyclic symbol prefix is be dded disrted by tking by or some trnsmitted smples from symbols. end Since of OFDM symbol uses period composite plcing IDFT m smples, t cyclic beginning. prefix is It dded provides by tking orthogonlity some smples between from subcrriers end of symbol by keeping period OFDM plcing symbol m t periodic beginning. in durtion It provides of orthogonlity extended symbol between for tht subcrriers reson, voiding by keeping intercrrier OFDM interchnnel symbol periodic interference in durtion simultneously. of extended In next symbol step, smpled for tht signl reson, is voiding converted intercrrier in n nlog wve interchnnel by digitl interference nlog simultneously. converter (DAC). In A furr next step, composite smpled wveform signl is is modulted converted in t n desired nlog wve RF for by trnsmission. digitl nlog The noticeble converter dvntge (DAC). A of furr OFDM composite over SC-FDM wveform is tht is modulted t domin desired representtion RF for trnsmission. of signls simplifies The noticeble signl dvntge error correction of OFDM t over receiver SC-FDM [33]. is tht domin representtion of signls simplifies signl error correction t receiver [33]. 4.2. SC-FDM 4.2. SC-FDM SC-FDM is hybrid modultion technique tht combines lloction flexibility multipth SC-FDM is resistnce hybrid modultion of OFDM technique or importnt tht combines chrcteristics of lloction single-crrier flexibility system. The multipth crucil resistnce chrcteristic of OFDM of SC-FDM or signl importnt genertion chrcteristics tht PAPR of single-crrier of finite system. shifted The signl crucil ends chrcteristic up being of sme SC-FDM s tht ofsignl originl genertion modulting is tht dt PAPR symbols of finite this is very different shifted from signl tht ends of OFDM up being for sme occupied s tht of bwidth originl in modulting sme dt dt rte. symbols However, if this chnnel is very bwidth different from is wider, tht of OFDM link between for sme symbol occupied length bwidth chnnel in bwidth sme dt cn rte. behowever, considered if s chnnel disdvntge bwidth of SC-FDM is wider, in comprison link between with OFDM symbol [33]. length Figure 6 presents chnnel bwidth block digrm cn for be considered signl processing s chin disdvntge of SC-FDM. of SC-FDM The first step in comprison smewith s for OFDM, [33]. modulting Figure 6 presents dt with block digrm for signl processing chin of SC-FDM. The first step is sme s for OFDM, modulting dt with one of modultion schemes for dt trnsmission over resource elements. The dt is plced over resource elements by djusting phse mplitude of

Electronics 2018, 7, 352 7 of 18 Electronics 2018, 7, x FOR PEER REVIEW 7 of 18 one of modultion schemes for dt trnsmission over resource elements. The dt is plced subcrrier over resource those elements derived for by djusting dt strem. phse Mmticlly, mplitude of it mens subcrrier multiplying those derived complex modultion for dtsymbol strem. Mmticlly, corresponding it mens subcrrier multiplying. complex After this modultion process, symbol re is nor block corresponding signl subcrrier processing. chin. After To convert this process, dt resymbols is norfrom block in time signl domin processing in chin. Todomin, convert dt DFT symbols is performed from fter time domin seril in prllel conversion. domin, As n DFT OFDM, is fterwrds, performed re fter is subcrrier seril prllel mpping conversion. n IDFT As n OFDM, trnsform fterwrds, signl re in is subcrrier mpping domin n IDFT trnsform signl in domin in time domin signl. The cyclic in time domin signl. The cyclic prefix cn be inserted when prllel seril conversion prefix cn be inserted when prllel seril conversion is done. Before modulting signl is done. Before modulting signl with high, pulse shping filter cn be used get with high, pulse shping filter cn be used get desired spectrum [34]. The originl desired spectrum [34]. The originl vlues of symbols cn be completely recovered if vlues of symbols cn be completely recovered if trnsmitted signl is properly smpled t trnsmitted signl is properly smpled t receiver. The lst two steps re sme s for OFDM. receiver. The lst two steps re sme s for OFDM. Thus, re is no difference in downlink Thus, re is no difference in downlink signl genertion chin. In sme wy, reverse of signl genertion chin. In sme wy, reverse of wht ws done t trnsmitter cn be wht ws done t trnsmitter cn be ccomplished t receiver. As mentioned in Section 4.1, ccomplished t receiver. As mentioned in Section 4.1, ech subcrrier crries only one prticulr ech modultion subcrrier symbol crries in OFDM, only one while prticulr DFT tkes modultion symbol symbol spreds in OFDM, it vi n while vilble DFT subcrrier tkes symbol in SC-FDM. spreds For tht it vi reson, n vilble SC-FDMsubcrrier is lso referred in SC-FDM. s DFT-spred For tht reson, OFDM SC-FDM [35]. Since is lso PAPR referred iss proportionl DFT-spred OFDM squre [35]. of Since number PAPR of subcrriers is proportionl involved, SC-FDM squre reduces of PAPR number by of subcrriers reducing involved, numbersc-fdm of subcrriers. reduces PAPR by reducing number of subcrriers. Figure Figure6. 6. Block Block digrm of SC-FDM modultion process. process. 5. Experiments Performnce Mesures 5. Experiments Performnce Mesures 5.1. Experiments 5.1. Experiments In generl, our work is divided in two prts. In first prt of our reserch, we hve performed In experimentl generl, our setup work by using is divided ArduPilot in two Meg prts. (APM) In 2.8 microcontroller first prt of our reserch, Mission Plnner, we hve performed which is suitble experimentl GCS ppliction setup by for using this microcontroller. ArduPilot The Meg APM (APM) 2.8 microcontroller 2.8 uses Mission micro ir Plnner, vehicle link which (MAVLink) is suitble procol [36] GCS ppliction mintin connection for this microcontroller. between APM 2.8 The APM GCS. 2.8 microcontroller During our experiments, uses micro APM ir 2.8 vehicle bordlink ws(mavlink) tested insted procol of drone [36] in order mintin obtin connection flight between dt tht were APM used 2.8 nlyze GCS. During SC-FDM our modultion experiments, for UAV APM communiction 2.8 bord ws system. tested insted As we of drone mentioned in order in Section obtin 3, flight UAV dt hs tht dtwere linkused tht utilizes nlyze SC-FDM rnge modultion from 150 for MHz UAV communiction system. As we mentioned in Section 3, UAV hs dt link tht utilizes rnge from 150 MHz 1.5 GHz. In our work, two 3DR 915 MHz telemetry rdios were used provide this dt link. To set up connection between APM 2.8 Mission Plnner GCS, one of 3DR 915 MHz telemetry rdios ws plugged in APM 2.8, s shown in Figure 7, while second telemetry rdio ws connected PC using USB cble. Since we hve communiction

Electronics 2018, 7, 352 8 of 18 1.5 GHz. In our work, two 3DR 915 MHz telemetry rdios were used provide this dt link. To set up connection between APM 2.8 Mission Plnner GCS, one of 3DR 915 MHz telemetry rdios ws plugged in APM 2.8, s shown in Figure 7, while second telemetry rdio ws connected Electronics 2018, 7, PC x FOR using PEER REVIEW USB cble. Since we hve communiction link between two telemetry 8 of 18 rdios, we cn check link sttus in terms of number of trnsmitted pckets, rnge, link between two telemetry rdios, we cn check link sttus in terms of number of trnsmitted number of chnnels, T x power, s shown in Figure 8. We configured connection settings using pckets, rnge, number of chnnels, T Mission Plnner. After estblishing successful connection, x power, s shown in Figure 8. We configured telemetry dt were obtined from connection settings using Mission Plnner. After estblishing successful connection, GCS. It should be noted tht se dt cn be represented s signls when y re crried by telemetry dt were obtined from GCS. It should be noted tht se dt cn be represented s MAVLink through wireless link between APM 2.8 GCS. All relevnt informtion bout signls when y re crried by MAVLink through wireless link between APM 2.8 APM 2.8 such s ltitude, yw, pitch, roll, ttitude, verticl speed, ground speed, etc., re shown in GCS. All relevnt informtion bout APM 2.8 such s ltitude, yw, pitch, roll, ttitude, verticl Flight speed, Dt ground screen speed, of etc., Mission re shown Plnner, in see Flight Figure Dt 8b. When screen we of move Mission APM Plnner, 2.8 microcontroller see Figure from 8b. When one plce we move nor, APM 2.8 Flight microcontroller Dt screen from will one disply plce nor, telemetry Flight dt ccording Dt screen new will position disply of APM telemetry 2.8. Afterwrds, dt ccording telemetry new dt position collected of APM from 2.8. Afterwrds, GCS were imported telemetry in dt MATLAB collected for from modultion GCS were process. imported in MATLAB for modultion process. Figure 7. Lyout of experimentl setup. () Figure 8. Cont.

Electronics 2018, 7, 352 9 of 18 Electronics 2018, 7, x FOR PEER REVIEW 9 of 18 (b) Figure 8. Mission Plnner ground control sttion (GCS): () Rdio link sttus; (b) Flight Dt screen. In second prt of of our our work, work, we we hve hve used used dt dt tht tht were were gred gred from from GCS GCS sttic simultion sttic simultion prmeters prmeters shown shown in Tblein 1Tble for 1 modultion for modultion process. process. For this prt, For this MATLAB prt, MATLAB ws used ws perform used perform modultion modultion process. process. Tble 1. Simultion prmeters. Prmeter Vlue System bwidth 10 10 MHz Frequency 2.4 2.4 GHz Modultion scheme BPSK, QPSK, 8-PSK, 16-QAM, 64-QAM Modultion BPSK, QPSK, 64-QAM Number of subcrriers 1024 Number Dt block of subcrriers size 1024 32 Dt Cyclic block prefix size 3264 Cyclic Pulseprefix shping filter 64 Rised-cosine (RC)/Root rised-cosine (RRC) Pulse Roll-off shping fcr filter Rised-cosine 0.3 (RC)/Root rised-cosine (RRC) Oversmpling fcr 5 Roll-off fcr 0.3 Number of itertions 10 4 Oversmpling Subcrrier spcing fcr 5 10 khz Number Equliztion of itertions 10Zero 4 forcing/minimum men squre error (MMSE) Subcrrier spcing 10 khz 5.2. Performnce Equliztion Mesures Zero forcing/minimum men squre error (MMSE) 5.2. Performnce The input dt Mesures symbols x n for 0 n N 1 re modulted by one of modultion schemes using n N-point DFT generte representtion of domin of input symbols n, The input SC-FDM dt output symbols sequence x n for 0X n N 1 re modulted one of modultion schemes k for 0 k N 1 is given by: using n N-point DFT generte representtion of domin of input symbols n, SC-FDM output sequence XN 1 k for 0 k N 1 is given by: j 2πnk X k = x n e N, 0 k N 1. (1) N 1 n=0 X k = x n e j2πnk N, 0 k N 1. (1) The SC-FDM symbol X[k] is complex n=0 number tht consists of rel imginry prts. According centrl limit orem, s number of subcrriers N gets lrger, rel The SC-FDM symbol X[k] is complex number tht consists of rel imginry prts. According centrl limit orem, s number of subcrriers N gets lrger, rel imginry prts of SC-FDM symbols follow norml (Gussin) distribution probbility density function (PDF) of X[k] cn be shown s [30]:

Electronics 2018, 7, 352 10 of 18 imginry prts of SC-FDM symbols follow norml (Gussin) distribution probbility density function (PDF) of X[k] cn be shown s [30]: where σ is strd devition. 5.2.1. PAPR f Xk (x) = x σ 2 e x2 2σ 2, (2) One of key prmeters for nlyzing performnce of trnsferred signl is its pek--verge power rtio (PAPR), which indictes how extreme peks re in wveform. This cn be determined s rtio of pek power verge power of trnsmitted signl [37]. In multicrrier system, PAPR occurs when different subcrriers do not correspond phse with ech or t ech point. It mens tht se subcrriers re different reltive ech or for different phse vlues. The vlue of PAPR depends on number of subcrriers involved s well s on modultion scheme. It should be mentioned tht high PAPR requires high power consumption for trnsmitting signl. In or words, it cn be sid tht efficiency of power mplifier will be very low. On or h, lower vlue of PAPR results in n increse in flight time (bttery life) of vehicle. The PAPR of trnsmitted signl is defined in units of db it cn be expressed s follows: ( ) X mx 2 PAPR db = 10 log 10, (3) X 2 ms where X mx is mximum vlue X ms is men squre vlue of signl. Here, PAPR is equivlent crest fcr, s it is defined in decibels. Now, if we consider tht X mx denotes crest fcr, it cn be written by: X mx = k=0,1,..., mx N 1 X k. (4) The cumultive distribution function (CDF) of X mx is probbility tht X mx will tke vlue less thn or equl x. The CDF of X mx is described by [38]: F Xmx (x) = P(X mx x) = x 0 f Y k (y)dy = x 0 y e y2 σ 2 2σ 2 dy, (5) CDF = 1 e x2 2σ 2. (6) Since we hve Equtions (5) (6), we cn chrcterize PAPR by using complementry cumultive distribution function (CCDF). The CCDF of PAPR is probbility of PAPR which is higher thn certin PAPR vlue it cn be clculted s: ) N CCDF = 1 P(PAPR x) = 1 (1 e x2 2σ 2. (7) 5.2.2. BER Anor importnt prmeter for mesuring performnce of wireless chnnel of UAV communiction system is bit error rte (BER). When dt is trnsmitted over wireless link between vehicle GCS, BER specifies number of errors tht pper in received dt. The environmentl conditions chnges propgtion pth re fundmentl resons for communiction chnnel degrdtion respective BER. To chieve n cceptble BER, i.e., for trnsmission of control comms on uplink, typiclly cceptble BER is round 10 6 10 9,

Electronics 2018, 7, 352 11 of 18 while tht cceptble vlue for trnsmission of pylod dt on downlink is 10 3 10 4 [39]. All vilble fcrs must be blnced. Usully, it is difficult chieve ll requirements some compromises re required. Moreover, higher level of error correction is needed in order recover originl dt. This cn help fix effects of ny occurred bit error. It results in fct tht overll BER cn be improved. If bidirectionl communiction between trnsmitter receiver is estblished very well signl--noise rtio (SNR) is high, n BER will be potentilly insignificnt will not hve n observble effect on whole UAV communiction system. The BER is equl number of bit errors (N E ) divided by tl number of trnsmitted bits (N T ), s expressed by following eqution: BER = N E N T. (8) The N E cn be computed by compring trnsmitted signl with received signl. The BER is most often expressed in terms of SNR. The SNR cn be defined s rtio of bit energy (E b ) noise power spectrl density (N o ), which is power per Hz it is expressed s follows: SNR = E b N o, (9) where E b is mesure of energy cn be defined by dividing crrier power by bit rte. The probbility of bit error (P b ) represents probbility tht error rte rises in received signl. The P b for M-ry PSK cn be defined s: [ P b 2E v π ) = 2Q sin( ], (10) N o M where E v is verge energy of trnsmitted symbol. Q represents scled form of complementry error function (erfc) it is given by: Q(x) = 1 2 er f c ( x 2 ). (11) It is necessry note tht ech different kind of modultion scheme hs its own vlue for error function. This is due fct tht ech type of modultion scheme executes in different wys in presence of noise. Finlly, we cn clculte P b for M-ry QAM by using following eqution: 5.2.3. Pulse Shping ( [ ] P b = 4 1 1 3E v )Q. (12) M (M 1)N o The pulse shping cn be used mke trnsmitted signl more suitble for communiction chnnel. In generl, pulse shping is importnt ensure correspondence of signl in its b. In this pper, rised-cosine (RC) root rised-cosine (RRC) pulse shping filters re used get desired spectrum. To perform response of RC filter, we use following eqution: T, ( )]} 0 f 1 α 2T T H RC ( f ) = 2 {1 + cos[ πt α f 1 α 1 α 2T, 2T f 1+α 2T 0, orwise (13)

Electronics 2018, 7, 352 12 of 18 where f is, T is symbol period, α is roll-off fcr its vlue cn be between 0 1. The representtion of time domin of this filter is given by: h RC (t) = sin( ) πt T cos παt [ T πt T 1 ( ) 2αt 2 ], (14) T where t is time. Equtions (13) (14) re used relize impulse responses of RC filter. Then, domin trnsfer function of RRC filter cn be written s: G RRC ( f ) = The impulse response of RRC filter is given by: H RC ( f ). (15) g RRC (t) = sin( πt T (1 α)) + 4αt T cos( πt T (1 + α)) [ ( ) ] 2. (16) πt T 1 4αt T Finlly, performnce of ech modultion scheme for SC-FDM OFDM cn be mesured by clculting different vlues of bove-mentioned prmeters. To chieve comprtive results tht re presented in following section, simultion prmeters dt tht were collected from GCS were utilized s inputs of system performed with MATLAB. 6. Results Anlysis In this reserch, MATLAB ws used perform PAPR simultions of SC-FDM OFDM s well s probbility of bit error simultion for SC-FDM using different kinds of modultion schemes such s BPSK, QPSK, 8-PSK, 16-QAM, 64-QAM. Of course, it is essentil select optiml number of subcrriers, types of pulse shping filters chnnel equliztion when simultion prmeters re inputted. The number of subcrriers symbols depends on cyclic prefix subcrrier spcing. When RC RRC pulse-shping filters re used filter symbol strem, y cn minimize n ISI. Hlf of this filtering cn be done on trnsmitter second hlf cn be done on receiver. The impulse responses of RC filter bsed on different roll-off fcrs hve been plotted shown in Figure 9. Furrmore, equlizer cn be used get recovery of trnsmit symbols by reducing n ISI. By removing ll ISI, zero forcing equlizer cn be optiml choice in noiseless chnnel. On or h, when chnnel is noisy, this equlizer significntly mplifies noise t frequencies. In this cse, minimum men squre error (MMSE) equlizer cn be more efficient thn zero forcing. The min function of n MMSE equlizer is minimizing ISI components entire power of noise in output insted of eliminting ISI completely. Figures 10 11 plot pek--verge power rtio (PAPR) mesurements of SC-FDM OFDM ginst complementry cumultive distribution function (CCDF). According simultion results, which re shown in Figures 10 11, SC-FDM hs n dvntge over OFDM becuse of its lower PAPR tht leds n increse in bttery performnce for UAV. As shown in Figure 10,b, it cn be seen tht PAPR vlue of SC-FDM for BPSK modultion scheme (5.6 db) is lmost sme s tht for QPSK (5.7 db) when CCDF is 10 4. At sme point, PAPR vlue of OFDM for BPSK is 8.8 db, for QPSK, it equls 8.6 db. In ddition, simultion results show tht 8-PSK hs slightly lower PAPR vlue thn 16-QAM for both SC-FDM OFDM. From Figure 11, it cn be observed tht incresing order of QAM modultion scheme (16-QAM 64-QAM) results in incresing PAPR vlues of SC-FDM OFDM from 8 db 10 db 10 db 12.2 db, respectively. From Figures 10 11, we cn conclude tht brupt chnge in CCDF vlue comes from fct tht PAPR vlues re expressed in logrithmic scle for both SC-FDM

hlf cn be done on receiver. The impulse responses of RC filter bsed on different roll-off fcrs hve been plotted shown in Figure 9. Furrmore, equlizer cn be used get recovery of trnsmit symbols by reducing n ISI. By removing ll ISI, zero forcing equlizer cn be optiml Electronics choice in 2018, 7, 352 noiseless chnnel. On or h, when chnnel is noisy, this equlizer 13 of 18 significntly mplifies noise t frequencies. In this cse, minimum men squre error (MMSE) equlizer cn be more efficient thn zero forcing. The min function of n MMSE equlizer is minimizing OFDM modultion ISI components techniques. The PAPR entire vlues power of of SC-FDM noise in OFDM output forinsted different of modultion eliminting schemes ISI completely. re presented in Tble 2. Electronics 2018, 7, x FOR PEER REVIEW 13 of 18 Figure 9. The impulse responses of remote control (RC) filter: () The impulse response; (b) response. Figures 10 11 plot pek--verge power rtio (PAPR) mesurements of SC-FDM OFDM ginst complementry cumultive distribution function (CCDF). According simultion results, which re shown in Figures 10 11, SC-FDM hs n dvntge over OFDM becuse of its lower PAPR tht leds n increse in bttery performnce for UAV. As shown in Figure 10,b, it cn be seen tht PAPR vlue of SC-FDM for BPSK modultion scheme (5.6 db) is lmost sme s tht for QPSK (5.7 db) when CCDF is 10 4. At sme point, PAPR vlue of OFDM for BPSK is 8.8 db, for QPSK, it equls 8.6 db. In ddition, simultion results show tht 8-PSK hs slightly lower PAPR vlue thn 16-QAM for both SC-FDM OFDM. From Figure 11, it cn be observed tht incresing order of QAM modultion scheme (16-QAM 64- QAM) results in incresing PAPR vlues of SC-FDM OFDM from 8 db 10 db 10 db 12.2 db, respectively. () From Figures 10 11, we cn conclude tht (b) brupt chnge in CCDF vlue comes from fct tht PAPR vlues re expressed in logrithmic scle for both SC- Figure 9. The impulse responses of remote control (RC) filter: () The impulse FDM OFDM modultion techniques. The PAPR vlues of SC-FDM OFDM for different response; (b) response. modultion schemes re presented in Tble 2. () (b) (c) Figure 10. The PAPR performncesof of SC-FDM OFDMfor for M-ry M-ry PSK PSK modultion schemes: () () BPSK; BPSK; (b) (b) QPSK; QPSK; (c) (c) 8-PSK. 8-PSK.

Electronics 2018, 7, x FOR PEER REVIEW 14 of 18 Electronics 2018, 7, 352 14 of 18 Electronics 2018, 7, x FOR PEER REVIEW 14 of 18 () (b) Figure 11. The pek--verge () power rtio (PAPR) performnces of SC-FDM (b) orthogonl division multiplexing (OFDM) for M-ry QAM modultion schemes: () 16-QAM; (b) 64- Figure 11. The pek--verge power power rtio rtio (PAPR) (PAPR) performnces performnces of SC-FDM of SC-FDM orthogonl orthogonl QAM. division multiplexing division multiplexing (OFDM) for(ofdm) M-ry QAM for M-ry modultion QAM modultion schemes: () schemes: 16-QAM; () (b) 16-QAM; 64-QAM. (b) 64- QAM. Tble 2. 2. PAPR results for SC-FDM OFDM. Tble 2. PAPR results for SC-FDM PAPR PAPR (db) OFDM. Modultion Modultion Scheme Scheme (db) SC-FDM SC-FDM OFDM PAPR (db) OFDM Modultion BPSK Scheme 5.6 8.8 BPSK SC-FDM 5.6 OFDM 8.8 QPSK QPSK 5.7 BPSK 5.6 5.7 8.6 8.8 8.6 8-PSK 8-PSK 7.6 9.7 9.7 QPSK 5.7 8.6 16-QAM 16-QAM 8 10 10 64-QAM 8-PSK 7.6 9.7 64-QAM 10 12.2 12.2 16-QAM 8 10 64-QAM 10 12.2 The probbility of bit error of different modultion schemes for SC-FDM is shown in Figure 12. From this figure, we cn observe tht BPSK QPSK modultion schemes chieve better The probbility of bit error of different modultion schemes for SC-FDM is shown in Figure 12. performnces thn or modultion schemes. It is evident tht BPSK QPSK modultion schemes From this figure, we cn observe tht BPSK QPSK modultion schemes chieve better re very suitble for SC-FDM modultion technique, ccording ir PAPR probbility performnces thn or modultion schemes. It is evident tht BPSK QPSK modultion schemes of bit error performnces, see Tble 2 Figure 12. re very suitble for SC-FDM modultion technique, ccording ir PAPR probbility of bit error performnces, see Tble 2 Figure 12. Figure 12. The probbility of bit error of of different modultion schemes for for SC-FDM. Figure 12. The probbility of bit error of different modultion schemes for SC-FDM. In our work, in order evlute complexity of proposed frmework in terms of computtion time, we used computer with withn n Intel(R) Core(TM) i7-4790 i7-4790 CPU CPU 3.60 3.60 GHz; GHz; RAM: RAM: 16.0 In our work, in order evlute complexity of proposed frmework in terms of 16.0 GB; Operting GB; Operting System: System: Windows Windows 8.1 Pro 8.1 64-bit Pro 64-bit MATLAB MATLAB R2016. R2016. As demonstrted As demonstrted in Figure 13, in computtion time, we used computer with n Intel(R) Core(TM) i7-4790 CPU 3.60 GHz; RAM: 16.0 GB; Operting System: Windows 8.1 Pro 64-bit MATLAB R2016. As demonstrted in Figure 13,

Electronics 2018, 7, 352 15 of 18 Electronics 2018, 7, x FOR PEER REVIEW 15 of 18 Figure computtion 13, computtion time of SC-FDM time of SC-FDM OFDM modultion OFDM modultion techniques techniques depends on depends number on of number subcrriers. of subcrriers. Figure 13. Computtion time comprison of SC-FDM OFDM. In Tble 3, we hve shown computtion time of SC-FDM OFDM modultion techniques. The computtion time of SC-FDM rnged from 222 6653 ms with N N [256, 4096], while rnge for OFDM ws between 350 9723 ms. We cn conclude tht computtion time of SC-FDM is lower thn tht of OFDM. Tble 3. Computtion time for SC-FDM OFDM. Tble 3. Computtion time for SC-FDM OFDM. Computtion Computtion Time Time (ms) Number of Subcrriers Number of Subcrriers SC-FDM OFDM 256 256 222 350 350 512 512 542 829 829 1024 1278 1916 1024 1278 1916 2048 2941 4348 4096 2048 6653 2941 4348 9723 4096 6653 9723 7. Conclusions 7. Conclusions In this pper, we hve nlyzed SC-FDM modultion technique in order provide n efficient In this pper, we hve nlyzed SC-FDM modultion technique in order provide n communiction link between vehicle GCS, since communiction link is n essentil prt efficient communiction link between vehicle GCS, since communiction link is n of UAV. The min purpose of this work ws nlyze SC-FDM modultion technique by essentil prt of UAV. The min purpose of this work ws nlyze SC-FDM modultion compring it with OFDM. In generl, this pper is divided in two prts. In first prt, UAV technique by compring it with OFDM. In generl, this pper is divided in two prts. In communiction link, UAV communiction system components, s well s some relted works hve first prt, UAV communiction link, UAV communiction system components, s well s some been discussed. On or h, second prt provided brief overview of SC-FDM relted works hve been discussed. On or h, second prt provided brief overview of OFDM modultion techniques discussed simultion results obtined by using vrious types of SC-FDM OFDM modultion techniques discussed simultion results obtined by using modultion schemes such s BPSK, QPSK, 8-PSK, 16-QAM, 64-QAM. The comprtive results vrious types of modultion schemes such s BPSK, QPSK, 8-PSK, 16-QAM, 64-QAM. The show tht SC-FDM is more effective thn OFDM for UAV communiction system, leding comprtive results show tht SC-FDM is more effective thn OFDM for UAV communiction noticeble improvement in terms of efficiency of UAV communiction link flight time of system, leding noticeble improvement in terms of efficiency of UAV communiction link qudcopters. Moreover, by nlyzing results of our work, we hve found tht BPSK QPSK re flight time of qudcopters. Moreover, by nlyzing results of our work, we hve found optiml modultion schemes for SC-FDM modultion technique. tht BPSK QPSK re optiml modultion schemes for SC-FDM modultion technique. Due fct tht currently used 2.4 GHz b for UAV communiction link is utilized Due fct tht currently used 2.4 GHz b for UAV communiction link is utilized by by mny different communiction systems leding genertion of interferences, some drones mny different communiction systems leding genertion of interferences, some drones hve hve lredy been designed operte in 5.8 GHz b. Thus, our furr reserch will focus on lredy been designed operte in 5.8 GHz b. Thus, our furr reserch will focus on nlyzing 5.8 GHz b for UAV communiction system. nlyzing 5.8 GHz b for UAV communiction system.

Electronics 2018, 7, 352 16 of 18 Author Contributions: Conceptuliztion, S.A. S.-H.L.; Softwre, S.A.; Forml Anlysis, S.-H.L. O.-H.K.; Writing-Review & Editing, S.A. O.-H.K.; Project Administrtion Funding Acquisition, K.-R.K. Funding: This reserch ws supported by MSIT (Ministry of Science ICT), Kore, under Gr Informtion Technology Reserch Center support progrm (IITP-2018-2016-0-00318) supervised by IITP (Institute for Informtion & communictions Technology Promotion), Bsic Science Reserch Progrm through Ntionl Reserch Foundtion of Kore (NRF) funded by Ministry of Science ICT (No. 2016R1D1A3B03931003, No. 2017R1A2B2012456) Kore Technology Informtion Promotion Agency for SMEs (TIPA) grnt funded by Kore government (Ministry of SMEs Strtups) (No. C0407372). Acknowledgments: The uthors would like thnk edirs reviewers for ir vluble time constructive comments. Conflicts of Interest: The uthors declre no conflict of interest. References 1. Ehichioy, D.; Golriz, A. Performnce comprison of OFDM DSSS on eronuticl chnnels. In Proceedings of 45th Interntionl Telemetering Conference, Ls Vegs, Nevd, USA, 26 29 Ocber 2009. 2. Mekw, K.; Negoro, S.; Tniguchi, I.; Tomiym, H. Power mesurement modeling of qudcopters on horizontl flight. In Proceedings of 5th Interntionl Symposium on Computing Networking (CANDAR 2017), Aomori, Jpn, 19 22 November 2017. 3. Sowh, R.A.; Acquh, M.A.; Ofoli, A.R.; Mills, G.A.; Koumdi, K.M. Rottionl energy hrvesting prolong flight durtion of qudcopters. IEEE Trns. Ind. Appl. 2017, 53, 4965 4972. [CrossRef] 4. Bosso, A.; Conficoni, C.; Tilli, A. Multiror UAV flight endurnce control: The drive perspective. In Proceedings of 42nd Annul Conference of IEEE Industril Electronics Society (IECON 2016), Florence, Itly, 23 26 Ocber 2016; pp. 1839 1845. 5. Kuzmenko, A.O. Optiml choice of technicl mens of UAV communiction link. In Proceedings of IEEE 3rd Interntionl Conference on Actul Problems of Unmnned Aeril Vehicles Developments (APUAVD 2015), Kyiv, Ukrine, 13 15 Ocber 2015; pp. 149 152. 6. Wu, Z.; Kumr, H.; Dvri, A. Performnce evlution of OFDM trnsmission in UAV wireless communiction. In Proceedings of 37th Soustern Symposium on System Theory (SSST 2005), Tuskegee, AL, USA, 20 22 Mrch 2005; pp. 6 10. 7. Miko, G.; Nemeth, A. SCFDM bsed communiction system for UAV pplictions. In Proceedings of 25th Interntionl Conference Rdioelektronik, Prdubice, Czech Republic, 21 22 April 2015; pp. 222 224. 8. Myung, H.G.; Lim, J.; Goodmn, D.J. Single crrier FDMA for uplink wireless trnsmission. IEEE Veh. Technol. Mg. 2006, 1, 30 38. [CrossRef] 9. Le, S.B.; Gwli, J.D.; Khrd, S.M. Performnce evolution of SC-FDMA for mobile communiction system. In Proceedings of 5th Interntionl Conference on Communiction Systems Network Technologies (CSNT 2015), Gwlior, Indi, 4 6 April 2015; pp. 416 420. 10. Thoms, P.A.; Mthurkni, M. Effects of different modultion schemes in PAPR reduction of SC-FDMA system for uplink communiction. Int. J. Adv. Res. Electron. Instrum. Eng. 2014, 3, 8531 8539. 11. Kur, N.; Gupt, N. Simultion nlysis of OFDM SC-FDMA with STBC using different modultion techniques. Int. J. Adv. Res. Comput. Eng. Tehnol. 2015, 4, 4184 4189. 12. Luo, Z.; Xiong, X. Performnce comprison of SC-FDMA-CDMA OFDM-CDMA systems for uplink. In Proceedings of Interntionl Conference on Consumer Electronics, Communictions Networks (CECNet 2011), Xinning, Chin, 16 18 April 2011; pp. 1475 1479. 13. Xiong, X.; Luo, Z. SC-FDMA-IDMA: A hybrid multiple ccess scheme for LTE uplink. In Proceedings of 7th Interntionl Conference on Wireless Communictions, Networking Mobile Computing (WiCOM 2011), Wuhn, Chin, 23 25 September 2011. 14. Myung, H.G. Introduction single crrier FDMA. In Proceedings of 15th Europen Signl Processing Conference (EUSIPCO 2007), Poznn, Pol, 3 7 September 2007; pp. 2144 2148. 15. Ahmed, J.J. PAPR reduction of loclized single crrier FDMA using prtil trnsmit sequence in LTE systems. Int. J. Comput. Newt. Techol. 2017, 5, 21 26. [CrossRef]

Electronics 2018, 7, 352 17 of 18 16. Rn, M.M.; Kim, J.; Cho, W.K. Performnce nlysis of subcrrier mpping in LTE uplink systems. In Proceedings of 9th Interntionl Conference on Opticl Internet (COIN 2010), Jeju, Kore, 11 14 July 2010. 17. Tsiropoulou, E.E.; Kpoukkis, A.; Ppvssiliou, S. Energy-efficient subcrrier lloction in SC-FDMA wireless networks bsed on multilterl model of brgining. In Proceedings of 2013 IFIP Networking Conference, Brooklyn, NY, USA, 22 24 My 2013; pp. 1 9. 18. Tsiropoulou, E.E.; Zirs, I.; Ppvssiliou, S. Service differentition resource lloction in SC-FDMA wireless networks through user-centric Distributed non-coopertive Multilterl Brgining. In Proceedings of Interntionl Conference on Ad Hoc Networks (ADHOCNETS 2015), Sn Remo, Itly, 1 2 September 2015; pp. 42 54. 19. Tsiropoulou, E.E.; Kpoukkis, A.; Ppvssiliou, S. Uplink resource lloction in SC-FDMA wireless networks: A survey txonomy. J. Comput. Netw. 2016, 96, 1 28. [CrossRef] 20. Luo, Z.; Xiong, X. Anlysis of effect of crrier offsets on performnce of SC-FDMA-IDMA systems. In Proceedings of 2nd Interntionl Conference on Consumer Electronics, Communictions Networks (CECNet 2012), Yichng, Chin, 21 23 April 2012; pp. 889 893. 21. Crespo, G.; River, G.G.; Grrido, J.; Ponticelli, R. Setup of communiction control systems of qudror type unmnned eril vehicle. In Proceedings of 29th Conference on Design of Circuits Integrted Systems (DCIS 2014), Mdrid, Spin, 26 28 November 2014. 22. Li, B. Study on modeling of communiction chnnel of UAV. J. Procedi Comput. Sci. 2017, 107, 550 557. [CrossRef] 23. M, J.; Liu, H. Performnce nlysis simultion of UAV dt link communiction link. In Proceedings of Interntionl Conference on Network, Communiction, Computer Engineering (NCCE 2018), Chongqing, Chin, 26 27 My 2018. 24. Li, J.; Ding, Y.; Fng, Z. Key techniques reserch on UAV dt link. J. Procedi Eng. 2015, 99, 1099 1107. [CrossRef] 25. Petrili, E.; Leccese, F.; Cini, L. Relibility mintennce nlysis of unmnned eril vehicles. Sensors 2018, 18, 3171. [CrossRef] [PubMed] 26. Lidbom, A.; Kiniklis, E. Providence UAV Support for Serch Rescue. Ph.D. Thesis, Deprtment of Signls Systems, Chlmers University of Technology, Gonburg, Sweden, 2015. 27. Mozffri, M.; Sd, W.; Bennis, M.; Debbh, M. Unmnned eril vehicle with underlid device--device communictions: Performnce trdeoffs. IEEE Trns. Wirel. Commun. 2016, 15, 3949 3963. [CrossRef] 28. Klntri, E.; Ynikomeroglu, H.; Yongcoglu, A. On number 3D plcement of drone bse sttions in wireless cellulr networks. In Proceedings of IEEE 84th Vehiculr Technology Conference (VTC-Fll 2016), Montrel, QC, Cnd, 18 21 September 2016. 29. Mozffri, M.; Sd, W.; Bennis, M.; Debbh, M. Efficient deployment of multiple unmnned eril vehicles for optiml wireless coverge. IEEE Commun. Lett. 2016, 20, 1647 1650. [CrossRef] 30. Rmvth, S.; Kshetrimyum, R.S. Anlyticl clcultions of CCDF for some common PAPR reduction techniques in OFDM systems. In Proceedings of IEEE Interntionl Conference on Communictions, Devices Intelligent Systems (CODIS 2012), Kolkt, Indi, 28 29 December 2012; pp. 405 408. 31. LTE Uplink SC-FDMA. Avilble online: https://www.exploregte.com/video.spx?video_id=55 (ccessed on 4 September 2018). 32. Kristensen, F.; Nilsson, P.; Olsson, A. A generic trnsmitter for wireless OFDM systems. In Proceedings of IEEE 14th Conference on Personl, Indoor Mobile Rdio Communictions (PIMRC 2003), Beijing, Chin, 7 10 September 2003; pp. 2234 2238. 33. Rumney, M. 3GPP LTE: Introducing Single-Crrier FDMA; Technicl Report; Agilent Technologies: Snt Clr, CA, USA, Jnury 2008. 34. Nigm, H.; Ptidr, M.K. Performnce evlution of CFO in single crrier-fdma. Int. J. Electr. Electron. Comput. Eng. 2014, 3, 104 110. 35. Girdhr, I.; Singh, C.; Kumr, A. Performnce nlysis of DFT spred OFDM systems. Int. J. Adv. Comput. Sci. Techol. 2013, 2, 21 26. 36. MAVLink Procol Overview. Avilble online: https://mvlink.io/en/procol/overview.html (ccessed on 15 September 2018).

Electronics 2018, 7, 352 18 of 18 37. Pervej, M.F.; Srkr, M.Z.I.; Roy, T.K.; Koli, M.N.Y. Impct nlysis of input output block size of DCT-SCFDMA system. In Proceedings of IEEE 17th Interntionl Conference on Computer Informtion Technology (ICCIT 2014), Dhk, Bngldesh, 22 23 December 2014; pp. 440 445. 38. Cho, Y.S.; Kim, J.; Yng, W.Y.; Kng, C.G. MIMO-OFDM Wireless Communictions with MATLAB; John Wiley & Sons: Singpore, 2010. 39. Biotti, S.; Sczzol, G.L.; Bttini, G.; Crovri, E. Advnces in UAV dt links: Anlysis of requirement evolution implictions of future equipment. In Proceedings of RTO SCI Symposium on Wrfre Aumtion: Procedures Techniques for Unmnned Vehicles, Ankr, Turkey, 26 28 April 1999. 2018 by uthors. Licensee MDPI, Bsel, Switzerl. This rticle is n open ccess rticle distributed under terms conditions of Cretive Commons Attribution (CC BY) license (http://cretivecommons.org/licenses/by/4.0/).