Visible Light Communications towards 5G

RADIOENGINEERING, VOL. 18, NO. 1, APRIL 2009 1

Visible Light Communications towards 5GStanislav ZVANOVEC 1, Petr CHVOJKA 1, Paul Anthony HAIGH 2, Zabih GHASSEMLOOY 3

1 Dept. of Electromagnetic Field, Czech Technical University in Prague, Technicka2, 166 27 Prague, Czech Republic

2 Faculty of Engineering, University of Bristol, Bristol, BS8 1TR, UK3 Optical Communications Research Group, Faculty of Engineering and Environment,

Northumbria University, Newcastle-upon-Tyne NE1 8ST, UK

[email protected], [email protected],[email protected], [email protected]

Abstract. 5G networks have to offer extremely highcapacity for novel streaming applications. One of themost promising approaches is to embed largenumbers of co-operating small cells into the macro-cell coverage area. Alternatively, optical wirelessbased technologies can be adopted as an alternativephysical layer offering higher data rates. Visible lightcommunications (VLC) is an emerging technology forfuture high capacity communication links (it has beenaccepted to 5GPP) in the visible range of theelectromagnetic spectrum (~370–780 nm) utilizinglight-emitting diodes (LEDs) simultaneously providedata transmission and room illumination. A majorchallenge in VLC is the LED modulation bandwidths,which are limited to a few MHz. However, myriadgigabit speed transmission links have already beendemonstrated. Non line-of-sight (NLOS) opticalwireless is resistant to blocking by people andobstacles and is capable of adapting its’ throughputaccording to the current channel state information.Concurrently, organic polymer LEDs (PLEDs) havebecome the focus of enormous attention for solid-state lighting applications due to their advantagesover conventional white LEDs such as ultra-low costs,low heating temperature, mechanical flexibility andlarge photoactive areas when produced with wetprocessing methods. This paper discussesdevelopment of such VLC links with a view toimplementing ubiquitous broadcasting networksfeaturing advanced modulation formats such asorthogonal frequency division multiplexing (OFDM) orcarrier-less amplitude and phase modulation (CAP) inconjunction with equalization techniques. Finally, thispaper will also summarize the results of the European

project ICT COST IC1101 OPTICWISE (Optical WirelessCommunications - An Emerging Technology) dealingVLC and OLEDs towards 5G networks.

Keywords5G networks, light emitting diodes,visible light communications

1. IntroductionIn recent years, the worldwide

growth in mobile data traffic has ledto the development of new technologiesfor future high capacity communicationsystems. Every year the number ofwireless devices such smartphones,laptops and tablets increases, thusmultimedia content becomes the mainpart of the overall mobile datatransferred. This fact results in anincreasing throughput requirement fromthe next generation of communicationnetworks (5G), which are expected to bedeployed beyond 2020. Network designersface several critical challenges, allof which need to be addressed, such asoptimal spectra allocation, highcapacity broadband links, powerconsumption, quality of services (QoS)and mobility. For instance,approximately one exabyte (EB) of datawas transferred across the entire

2 A. SURNAME, C. SECONDNAME, EXEMPLARY DOCUMENT FOR THE PAPER IN RADIOENGINEERING

global internet in 2000 (0.083EB/month) [1]. In contrast, ~30 timesmore data was carried by the mobilenetworks per month in 2014, whichcorresponds to ~2.5 EB/month (refer toFig. 1). Moreover, the latestprojections from Cisco predict that theoverall mobile data traffic will reachapproximately 24 EB/month by 2019,which is approximately one order ofmagnitude larger than 2014 (Fig. 1).This corresponds to a compound annualgrowth rate (CAGR) of 57% for the 2014-2019 period. Following the projectionFig. 1 also shows the fit for thisdata, which estimates that > 30EB/month will be transmitted beyond2020. Most of this mobile data traffic(up to 69%) is expected to consist ofvideo and media by the end of 2018 [1].

Fig. 1. A prediction of the mobile datatraffic per month

5G networks are expected to meet allthe mentioned requirements [2], [3].The network architecture is expected tobe changed dramatically and the limitedfrequency bands must be used moreefficiently. Future systems will bebased on heterogeneous networks(HetNets) and advanced radio accesstechnologies (RATs). HetNets includeseveral small cells featuring lowtransmission power and small coveragearea, thus enabling high cell density.Such a system configuration allowsspectral reuse, hence improving thecapacity of the wireless channel [2],[3], [4]. The cellular architectureshould be designed to separate indoor

and outdoor scenarios and supportmassive multiple-input multiple-output(MIMO) technologies with distributedantenna systems [2], [3].

As the number of communicationdevices increases and the demand forconnections grows, power efficiencybecomes one of the most importantissues for 5G networks. Around 2% ofall carbon-dioxide emissions worldwideare produced by communicationtechnologies; this will increasesignificantly with the aforementionedincreasing mobile data traffic demand[5]. Moreover, ~57% of overall wirelessnetwork energy consumption isdissipated in radio access nodes [6].Thus, ‘green’ and energy efficientsystems should reduce the CO2 emissionsand decrease the operating costs.

Indoor communication systems canoffer a solution to all the aboveissues, such as mm-wave systems (3-300GHz) or optical wireless communicationssuch as VLC, which is carried at 400 –490 THz. 5G-VLC offers a number ofsmall cells, also referred to asattocells in the literature, in theindoor environment, thus enabling manyadvantages such as high capacity datatransmission, excellent mobility andenergy efficient management. Recentexperiments demonstrated up to1.6 Tbit/s for optical wirelessbackhaul links at 1550 nm [7] and up to3.4 Gbit/s [8] in VLC based accessnetworks. Moreover, two functionalitiesare provided with VLC, i.e. thecombining of data transmission withillumination of the room using light-emitting diodes (LEDs), which is notavailable with other networktechnologies. This paper provides awide overview of the VLC technology andsummarizes its development and thestate-of-the-art. The rest of the paperis organized as follows: Section 2describes the fundamental VLCprinciples with emphasis on inorganicand organic LEDs, Section 3 is focused


on the modulation formats and finallyin Section 4, perspective VLCapplications are discussed.

2. Visible Light Communications TechnologyLicense-free spectrum, practically

unlimited channel bandwidth, highcapacity links and energy efficiencyare the main features of light fidelity(Li-Fi) networks. Li-Fi is a subset ofthe VLC domain, which refers tobroadcasting access networks withmultiple users. The transmitterconsists of LEDs either singularly orin an array that are intensitymodulated at a rate above which it isnot perceivable by the human eye. Sincefull room illumination is required, thevisible range of the electromagneticspectrum (~370 – 780 nm) is utilized inVLC. Moreover, VLC has been proposedfor future 5G networks standards.

There are two type of LEDs used toproduce white light: (i) inorganic metalalloy semiconductor LEDs (usually ablue emitting gallium nitride (GaN)interface with a cerium doped yttriumaluminium garnet (Ce:YAG) colourconverting phosphor) and (ii) organicLEDs (OLEDs) made from either smallmolecules or polymers, using epitaxialor wet processing methods, respectively[9], [10]. Inorganic LEDs are commonlyused in VLC, while generic OLEDs areattracting significant attention forfuture VLC networks. When dissolvedinto solvents and processed with wetmethods (i.e. inkjet printing, spraycoating) polymer based OLEDs offerseveral significant advantages overinorganic devices, namely; large,arbitrarily shaped photoactive areaslimited only by the size of theprinting apparatus, mechanicalflexibility, low temperature andultralow costs.

2.1 Inorganic LEDsThe most common type of inorganic

LEDs producing white light are Ce:YAGconverted GaN diodes, as mentioned,which are known as white phosphor LEDs(WPLEDs). Alternatively, white lightcan be produced using a single LEDpackage with on-board red, green andblue (RGB) chips. Considering WPLEDs,the main impediment in achieving highdata rates is the modulation bandwidth.GaN diodes can offer modulationbandwidths up to several hundred MHz[11], however the Ce:YAG phosphor layerhas a slow transient response, reducingthe bandwidth down to the low MHzregion, which common values around 4-5MHz [12]. Thus, increasing thetransmission capacity is the keychallenge undertaken by researchers.However, the concept of micro-LEDs wasintroduced in [13], where thephotoactive area of the device isreduced significantly to the µm scale,thus relieving the plate capacitance ofthe device and increasing thebandwidth. Depending on the photoactivearea diameter, bandwidths exceeding 400MHz were demonstrated. Such a devicecan provide a 3.22 Gb/s throughput whenusing an adaptive bit- and power-loading technique [14]. On the otherhand, reducing the photoactive areameans a reduction in optical power,which in turn severely limits thetransmission distance. In [13], theoptical power ranged from ~0.5 mW to ~5mW, which is significantly smaller thanstandard GaN LEDs.

VLC links at Gb/s data rates havealready been demonstrated, despite thevery limited LED modulation bandwidths.A popular method to increase thecapacity of VLC links is to usespectrally efficiency modulationformats such as discrete multi-tone(DMT) [8, 15]. A 3.4 Gb/s transmissionspeed was achieved in [8] using asingle RGB LED and DMT modulation at a

Fig. 2. The structure of OLED


distance < 30 cm. Wavelength divisionmultiplexing (WDM) was utilized totransmit independent streams ofinformation on each wavelength.Nevertheless, DMT requires complexsignal processing and a feedbackchannel for bit- and power-loading. Asimilar approach was adopted in [15]resulting in a transmission speed of 1Gb/s at a 10 cm distance.

Another possibility for increasingthe channel capacity is theimplementation of an equalizer [12].On-off keying (OOK) is the most commonmodulation format due to simplicity ofimplementation and compatibility withequalizers. In [12] several equalizerswere tested on a commercially availabledigital signal processing (DSP) board.The LED had low modulation bandwidth(4.5 MHz) and transmission speeds up to170 Mb/s were recorded using a highlycomplex artificial neural network (ANN)based equalizer. Real time dataprocessing on field programmable gatearrays (FPGAs) is the next step towardsbuilding a fully real time system, ascurrently most reports in theliterature feature offline processing.

2.2 Organic LEDsThere are two types of OLEDs, the

first produced using small molecules(SMOLEDs) and the second with polymers(PLEDs). Due to the previouslymentioned advantages, organic materialshave been focus of the research to beimplemented in communication systems.According to [16], the global organicelectronic market will approach$80 billion by 2020, thus resulting in(CAGR) of 29.5% between 2014 – 2020.

A possible and generic PLEDstructure is depicted in Error:Reference source not found. It consistsof thin films (total thickness 1-200nm) of organic emissive and chargetransport layers. Polymers such as:

(i) red: poly[2-methoxy-5-(3′,7′-dimethyloctyloxy)-1,4-phenylenevinylene] (MDMO-PPV)

(ii) green: poly[(9,9-di-n-octylfluorenyl-2,7-diyl)-alt-(benzo[2,1,3]thiadiazol-4,8-diyl)] (F8BT) and

(iii) blue: poly(9,9-dioctylfluorene), (poly(9,9-dioctylfluorene-alt-N-(4-butylphenyl)diphenylamine) andpoly(9,90-dioctylfluorene-alt-bis-N,N0-(4-butylphenyl)-bis-N,N0-phenyl-1,4-phenylenediamine) (F8:TFB:PFB)

are evaluated as the emissive layer in[17] while a poly(3,4-ethylenedioxythiophene) (PEDOT) dopedwith poly(styrenesulfonic acid) (PSS)interlayer is used consistently for thecharge transport layer in order tominimize the energy gap between theelectrode and emissive layer. In Error:Reference source not found the PLED isa bottom emitter and hence the anodemust be transparent (generally indiumtin oxide (ITO) is used). When positiveand negative charges recombine, thecharges combine for a fraction of asecond as Frenkel excitons before theenergy is emitted as photons. Thematerial can be placed on varioussubstrates including plastic, enablingflexibility.

However, the most limiting factorfor OLEDs is very low modulationbandwidth, which is much lower than thebandwidth of conventional inorganicdiodes. The reason for this is becausethe charge transport characteristics of


organic materials are approximatelythree orders of magnitude lower thanthose of amorphous silicon (a-Si).OLEDs act as a low pass filter with acertain cut-off frequency. The -3 dBfrequency is given by:

fc=1

2πRC(1)

where R is the effective resistance andC is the plate capacitance, given as:

C=Aϵ0ϵr/d(2)

where A is the photoactive area and d isthickness of OLED, ε0 and εr are therelative permittivity of free space andthe emissive layer, respectively. As dis generally very small (1-200 nm),this tends to a large C, hence a veryhigh capacitance and low bandwidth,which is typically in the order ofseveral hundred kHz [18]. On the otherhand, OLEDs with bandwidths up to ~60MHz were reported in [19], which wasachieved by reducing the photoactivearea to 0.018 mm2. Using commercial andcustom OLEDs, promising experimentalresults have been reported in theliterature. A SMOLED with a bandwidth93 kHz is modulated by DMT signal todemonstrate to data rate up to 1.4 Mb/sin [20]. A two-fold improvement of thistransmission speed was published in[21] using the same SMOLED as in [20];a 2.7 Mb/s link is shown using a multi-layer perception (MLP) ANN equalizeroffline in MATLAB.

A breakthrough in organic VLC wasmade in [22], where PLEDs were used todemonstrate transmission speeds inexcess of that required for Ethernetconnectivity. The data rate of 20 Mb/swas reported in [22, 23] using OOKformat and an MLP equalizer. Morerecently in [17], an aggregatedtransmission speed of 55 Mb/s wasexperimentally demonstrated and wasachieved using an RGB PLED and WDM,

which is a promising improvement forfuture research activities.

2.3 Other organic componentsOrganic photodetectors (OPD) are

also a promising technology and cansubstitute silicon photodiodes inspecific applications. OPDs used in theauthors’ previous work are based on thebulk heterojunction principle [24]; aninterpenetrated and disordered blend ofan electron donor and electron acceptorand were fabricated using spray coating[25]. The material costs for thepoly(3-hexylthiophene):[6,6]-phenylC61-butyric acid methylester(P3HT:PCBM) blend are around ~€0.20/cm2.Furthermore, OPDs can offer superiorresponsivity in comparison to Siphotodetectors in the visible spectrumas shown in Fig. 2.

Fig. 2 Photodetector material responsivities

The OPD with raw 3 dB bandwidth of30 kHz is tested in [26] with theresulting data rate of 750 kb/s, whichis ~20 times increment representinghuge potential of such devices. An OPDwith ~160 kHz modulation bandwidth isutilized in [27]. Fourth-order pulseposition modulation (4-PPM) and an ANNequalizer were used to achieve atransmission speed of 3.75 Mb/s. Fullyorganic VLC link was introduced in[28], where the low bandwidthcomponents (up to 135 kHz) with OOKmodulation and ANN based filtering were


used to achieve data rate exceeding 1Mb/s.

3. Modulation formatsBesides equalization techniques,

spectrally efficient modulation formatssuch as orthogonal frequency divisionmultiplexing (OFDM) are a popular wayto increase the VLC link capacity. Forinstance, Gb/s transmissions werereported in [8] and [29] byimplementing DMT. Power efficientschemes such as low peak-to-averagepower ratio (PAPR) single carrierfrequency-domain equalization (SC-FDE)was proposed in [30]. Here the IFFTblock was moved to the receiver side toavoid the generation of complicatedwaveforms and consequently reducedPAPR. Further spectral efficiencyimprovements have been achieved byadopting novel OFDM SC-FDE signalformats - polar OFDM and polar SC-FDEin [31].

Recently, carrier-less amplitude andphase modulation (CAP) has appeared asa candidate for optical systems, whichoutperforms OFDM channel using the sameexperimental setup [32]. CAP systemshave several advantages over OFDMincluding no Fourier transform as inOFDM or local oscillator, which isutilized in a single carriermodulations such as quadratureamplitude modulation (QAM). The carrierfrequencies in CAP are generated byeither analogue or digital finiteimpulse response filters (FIR).

Unlike OFDM, a flat frequencyresponse is required for CAP meaningthe low bandwidths available are themajor problem once more, due to the 20dB/decade attenuation for frequenciesoutside the bandwidth. A possiblesolution for this was introduced in[33], where the available bandwidth isdivided into a multiband (m-CAP) formatfor an optical fiber channel. The

transmission bandwidth was split into 6sub-bands (subcarriers) and theperformance was compared with thetraditional CAP (1-CAP) system. Thefibre system transmission speed was102.4 Gb/s and 100 Gb/s for m-CAP and1-CAP, respectively, which is not asignificant gain, however a gain intransmission speed was not the focus ofthe article. The m-CAP system noted asignificant improvement in dispersiontolerance. Nevertheless, for 1-CAPtransmission, two FIR filters arerequired for signal generation. As mincreases, the number of FIR filtersgrows by 2m. Thus, one must consider anincrement in the system complexityresulting from a multiband approach. Onthe other hand, such a concept allowsoptimization of the modulation formatused in each sub-band according tomeasured signal-to-noise ratio (SNR),i.e. bit- and power-loading in OFDM.The principle of m-CAP is illustratedin Fig. 3. The available systembandwidth is split into m sub-bands.The more sub-bands are utilized, theless bandwidth is occupied by a singlesubcarrier, thus resulting in a reducedattenuation caused by the LED low-passfrequency response.

Fig. 3. The concept of m-CAP modulation,where the available system bandwidth is splitinto 1, 4 and 10 sub-bands


A detailed description of the m-CAPmodulation format and signal generationcan be found in [34, 35]. The first VLCexperiment utilizing m-CAP was reportedin [35], where a data rate of 31.5 Mb/swas demonstrated using 10-CAP;resulting in a spectral efficiency of4.85 b/s/Hz, which offers hugepotential for a future research.

4. ApplicationsAs the technology has evolved over

the last decade and with rapid increaseof mobile data requirements, there areseveral challenging areas for specificdeployment of VLC systems. Alongsideclassic indoor communication schemes(i.e., broadcasting networks), VLC canalso be utilized for localization orcar-to-car and car-to-infrastructurecommunications. The following sectionshighlight the main approaches andprinciples.

4.1 Indoor communications

Alongside illumination and datacommunications, VLC systems have beenproposed for indoor positioning with avery high accuracy (a few cm) [x]. TheIEEE 802.15.7 task group [36] has beenforming the new standard for VLC,orienting their efforts towards the PHY

and MAC standards since 2009. The mainfocus from the first releases had beengiven on slow VLC for indoorpositioning by means of optical cameracommunication. More recently high bitrate VLC transmission systems are understandardization, especially based onresults produced by the EU COST IC1101project OPTICWISE consortium.

For indoor applications with a staticenvironment, VLC can offer high QoSeven when the movement of people andshadowing by obstacles perturbs thebeam (so called non-line of sight(NLOS)). This has been proved byseveral theoretical and experimentalinvestigations [37], [38]. Typicallysuch NLOS systems use diffusereflections, where the transmitterilluminates the ceiling, and/or widefield-of-view or receiving aperturesinstead of direct links. Data rates upto 400 Mbit/s have been reported for aNLOS link in [39]. The first mobile VLCsystem was reported for typical indoordistances between 2m – 20m with datarates decreasing with distance fromapprox. 500 Mbit/s to 100 Mbit/s in[40].

The typical model of diffusereflection was described in [41]. Thepower efficiency for the diffuse signalcan be derived as [41]:

μdiff=PdiffPT

=ARsin

2 (FOV /2 )Aroom

⟨ρ⟩1−⟨ρ ⟩

(3)

where ⟨ρ ⟩ is mean reflectivity, FOVstands for field of view of detectorand ARand Aroom are areas of detectorand room, respectively. The decay timethen can be given as [41]:

τ=−⟨t ⟩ln ⟨ρ ⟩

(4)


The average time ⟨t ⟩ between twodiffuse reflections for a rectangularroom with dimensions l × w × h (length ×width × height) is expressed by [41]:

⟨t ⟩= 2lwhc (lw+lh+wh)

(5)

The DC gain of the reflected path canafterwards be determined as Error:Reference source not found:

Href (O )={(m+1 )

2 (πd1d2 )2ρAdetdAwcos

m (θr )cos (α )∙

∙cos (β )Ts (ϑ )g (ϑ )cos (ϑ ),for0≤ϑ≤ϑFOV0forϑ>ϑFOV

(6)

where d1 is the distance betweentransmitter and the reflective point, d2is the distance between the reflectivepoint and the receiver, ρ is thereflectance coefficient, dAw is a smallreflective area on the wall, α is theangle of incidence from the transmitterand β is the angle of irradiance from areflected point.

Fig. 4 illustrates people’s movementwithin a typical office environmentutilizing VLC by employing 4 LEDsmounted on the ceiling.

Fig. 4. The simulation of people’s movementwithin an office using VLC with 4-LED

In such a case sometimes the directpaths from transmitter to receiver areblocked or temporally shadowed by ahuman. Several studies derived thepercentage of shadowing. For instance[42] reported shadowing with aprobability of < 2% for a multiple-input multiple-output (MIMO) systemcovering a typical office. Similarresults were derived in [43] where timedivision multiple access wasinvestigated. Higher order reflectionsinduce significant influence on thetemporal dispersion according to [37],[38]. The reflection component alwaysappears respective to LOS incidentpaths [38]. This influence can beeasily observed in the transmissionbandwidth. Two examples of a normalisedchannel response for the scenario shownin Fig. 4 are illustrated in Fig. 5.

Fig. 5. Examples of the normalized channelimpulse response for LOS/NLOS scenarioconsidering blocking signal path by people in anoffice with: a) 4 LEDs, b) 18 LEDs

The measured probability densityfunction of the normalized receivedoptical power considering people’s


movement as well as the analysis of theRMS delay spread for different indoorscenarios and people densities in roomswas carried out in [44]. Based on themeasurement campaign the normalizedreceived power showed a Rayleighdistribution with the scale parametervarying from 0.98 to 1.79 for an emptyto a crowded room. The RMS delay spreadstatistics have been derived for threedifferent indoor scenarios. For thecase of furnished office environment(people density > 0.16 people/m2), thecumulative distribution function (CDF)of the received power differs in theworst case by up to 7% contrary to anRMS delay of 2 % that was experiencedunder the same people density in thecorridor [44].

4.2 Positioning and localization

Several positioning systems havebeen tested over the last few years. In[45] a digital camera was used as areceiver to capture a sequence ofimages of the LED positioning beacontransmitter. By using image-processingalgorithms, the system was able todecode the location information encodedin the visual patterns transmitted byLEDs. The system demonstrated thatimproved performance can be attainedeven at low values of SNR. A system forlocalization of vehicles using theglobal positioning system (GPS)together with a light beacon devicemounted on the vehicle to receiveinformation from transmitter positionedat the road intersections is developedby Honda motors Co., Ltd. in 2010[46].

Fig. 7. Concept of VLC positioning system(multi-access mechanism among single terminal andmultiple lights + position estimation)

Another challenge represents theutilization of the localization withinthe indoor scenario, where standard GPSsignals cannot be received (seeillustrative deployment in Fig. 7). Inrecent years, we have seen research anddevelopment in optical based indoorpositioning schemes (IPS) offersmultitude of advantageous includingsmaller transceiver size, immunity toelectromagnetic interference, andinherent security [47]. VLC based IPSoffers the advantage of LED and VLCtechnologies such as ubiquitouscoverage, static channel, multiplelighting elements etc. [48]. Using VLCwith synchronization between thetransmitter and receiver, the bounds onposition estimation accuracy aretypically in the order of millimetersor centimeters depending on thegeometry of the room, the frequency andpower of the transmitted signal and theproperties of the LED and thephotoreceiver [49]. A VLC-based IPSemploying 3-LED, the dual-tone multi-frequency technique and a dedicatedalgorithm was reported in [48]. Unlikethe time-division multiplexing basedVLC-IPS, this scheme does not requiresynchronization between the transmitterand receiver, thus makes it simple,robust and cost effective. VLC-IPS ishighly accurate offering an averagepositioning error of about 1.6 cm,which much less than many existing IPS.


Typically several transmitters serveas beacons and the RMS delay spread canbe used to quantify the amount ofmultipath distortion that can occur ata particular point within a room [50].Parameter DrmsMax corresponds to thelargest multipath distortion, whichlimits the maximum transmission datarate Rmax of the system. Rmax for theindoor VLC channel can be calculatedfollowing [51], which is given by:

Rmax≤1

10DrmsMax

.(7)

In order to evaluate the positioningaccuracy, we have to consider theCramer-Rao bound (CRB) [52] as aperformance reference, which is thelower bound on the mean squareestimation error in the set of unbiasedestimates. Typical CRB ranges withinthe rooms for 4-, 6- and 9-cellconfigurations were derived for anoptimized Lambertian order (OLO) LEDcase for an indoor cellular opticalwireless communication system in [53].Simulations of particular scenariorevealed reached values of CRB from12.8 cm, from 8.6 cm and from 5.8 cm,respectively, for above mentionedcells’ deployment in for rooms of 5 × 5× 3 m3, 4 × 6 × 3 m3, and 5 × 5 × 3 m3

[53]. This has shown VLC as very usefultool for developers of indoorpositioning systems.

4.3 Car-to-car communicationsVLC can also be utilized for outdoor

applications such as the publictransport. Note how the infrastructureof public lights have changes over thelast 5 years. Typical incandescentlamps have been replaced by LEDlighting across whole cities. Forexample the Los Angeles LED StreetlightReplacement Program has replaced over140,000 existing streetlights in the

city with LED units which has broughtenergy saving 68 GWh/year and moneysaving $10M/year [54].

Several test use-cases andexperimental results have beenpublished for a vehicular VLC networkconsisting of on-board units, vehicles,and road side units, i.e., trafficlights, street lamps, digital signageetc. Carss fitted with LED-based frontand back lights can communicate witheach other and with the road side units(RSUs) through the VLC technology.Furthermore, LED-based RSUs can be usedfor both signaling and broadcastingsafety-related information to vehicleson the road. Such a network isillustrated in Fig. 8.

Fig. 8. Concept of VLC for car to car and carto infrastructure

An analytical performance analysisof VLC based car-to-car communicationsfor a range of communication geometriesconsider both the LOS and NLOS pathsover a link span of 20 m at a datarate of 2 Mbps was outlined in [55].Optical wireless communications systemsbased on an LED transmitter and camerareceiver was proposed for automotiveapplications in [56]. The signalreception experiment has been performedfor static and moving camera receivers.Up to 15 Mb/s error-free throughputunder fixed conditions was sustained.This represents very good performance


for optical wireless communicationsystems, since the experiment did notinvolve further correction methods likecoding and equalization. In [57] it wasshown that the receiver in the drivingsituation can detect and accuratelytrack an LED transmitter array witherror-free communication over distancesof 25 – 80m. Further tests andexperiments are however needed to provethese concepts.

5. ConclusionThe research and development in VLC

at a global level has increased morethan ten times for last two years. Thispaper gave an overview of VLCs and itsuse in a number of applications. Inoutdoor environment VLC can provideinternet hot spots using streetlighting and mobile access as part ofthe 5G technology in highly congestedareas and within indoor environment itcan be used for localization and smallcells coverage networks. VLC has beenaccepted as part of the 802.15.7 taskgroup and is being proposed as asupplement technology in 5G networks.We assume further increase researchefforts within selected applications.

AcknowledgementsThis joint research is supported by theEU COST ICT Action IC1101 Optical Wireless Communications: An Emerging Technology (OPTICWISE) and by the GrantAgency of the Czech Technical University in Prague, grant no. SGS14/190/OHK3/3T/13.

References[1] C. V. N. Index, "Global Mobile Data

Traffic Forecast Update, 2014-2019, Whitepaper," ed, 2013.

[2] B. Bangerter, S. Talwar, R. Arefi, and K.Stewart, "Networks and devices for the 5G

era," Communications Magazine, IEEE, vol. 52,pp. 90-96, 2014.

[3] W. Cheng-Xiang, F. Haider, G. Xiqi, Y.Xiao-Hu, Y. Yang, Y. Dongfeng, et al.,"Cellular architecture and keytechnologies for 5G wireless communicationnetworks," Communications Magazine, IEEE, vol.52, pp. 122-130, 2014.

[4] Q. C. Li, N. Huaning, A. T.Papathanassiou, and W. Geng, "5G NetworkCapacity: Key Elements and Technologies,"Vehicular Technology Magazine, IEEE, vol. 9, pp.71-78, 2014.

[5] S. Yong Sheng, T. Q. S. Quek, M.Kountouris, and S. Hyundong, "EnergyEfficient Heterogeneous CellularNetworks," Selected Areas in Communications, IEEEJournal on, vol. 31, pp. 840-850, 2013.

[6] R. Q. Hu and Q. Yi, "An energy efficientand spectrum efficient wirelessheterogeneous network framework for 5Gsystems," Communications Magazine, IEEE, vol.52, pp. 94-101, 2014.

[7] G. Parca, A. Shahpari, V. Carrozzo, G. M.Tosi Beleffi, and A. L. J. Teixeira,"Optical wireless transmission at 1.6-Tbit/s (16×100 Gbit/s) for next-generation convergent urbaninfrastructures," Optical Engineering, vol. 52,pp. 116102-116102, 2013.

[8] G. Cossu, A. M. Khalid, P. Choudhury, R.Corsini, and E. Ciaramella, "3.4 Gbit/svisible optical wireless transmissionbased on RGB LED," Optics Express, vol. 20,pp. B501-B506, 2012/12/10 2012.

[9] J. H. Burroughes, D. D. C. Bradley, A. R.Brown, R. N. Marks, K. Mackay, R. H.Friend, et al., "Light-emitting diodes basedon conjugated polymers," Nature, vol. 347,pp. 539-541, 10/11/print 1990.

[10] C. W. Tang and S. A. VanSlyke, "Organicelectroluminescent diodes," Applied PhysicsLetters, vol. 51, pp. 913-915, 1987.

[11] J. J. D. McKendry, R. P. Green, A. E.Kelly, G. Zheng, B. Guilhabert, D.Massoubre, et al., "High-Speed Visible LightCommunications Using Individual Pixels ina Micro Light-Emitting Diode Array,"Photonics Technology Letters, IEEE, vol. 22, pp.1346-1348, 2010.

[12] P. A. Haigh, Z. Ghassemlooy, S.Rajbhandari, I. Papakonstantinou, and W.Popoola, "Visible Light Communications:170 Mb/s Using an Artificial NeuralNetwork Equalizer in a Low Bandwidth WhiteLight Configuration," Lightwave Technology,Journal of, vol. 32, pp. 1807-1813, 2014.


[13] J. J. D. McKendry, D. Massoubre, S. Zhang,B. R. Rae, R. P. Green, E. Gu, et al.,"Visible-Light Communications Using aCMOS-Controlled Micro-Light- Emitting-Diode Array," Lightwave Technology, Journal of,vol. 30, pp. 61-67, 2012.

[14] D. Tsonev, C. Hyunchae, S. Rajbhandari, J.J. D. McKendry, S. Videv, E. Gu, et al., "A3-Gb/s Single-LED OFDM-Based Wireless VLCLink Using a Gallium Nitride μLED,"Photonics Technology Letters, IEEE, vol. 26, pp.637-640, 2014.

[15] A. M. Khalid, G. Cossu, R. Corsini, P.Choudhury, and E. Ciaramella, "1-Gb/sTransmission Over a Phosphorescent WhiteLED by Using Rate-Adaptive DiscreteMultitone Modulation," Photonics Journal, IEEE,vol. 4, pp. 1465-1473, 2012.

[16] R. Singh, Global Organic Electronics Market(Application and Geography) - Size, Share, Global Trends,Company Profiles, Demand, Insights, Analysis, Research,Report, Opportunities, Segmentation and Forecast, 2013– 2020, 2014.

[17] P. A. Haigh, F. Bausi, H. Le Minh, I.Papakonstantinou, W. Popoola, A. Burton, etal., "Wavelength-Multiplexed Polymer LEDs:Towards 55 Mb/s Organic Visible LightCommunications," IEEE Journal on Selected Areas inCommunications, vol. [Accepted], 2014.

[18] P. A. Haigh, Z. Ghassemlooy, S.Rajbhandari, and I. Papakonstantinou,"Visible light communications usingorganic light emitting diodes,"Communications Magazine, IEEE, vol. 51, pp. 148-154, 2013.

[19] I. A. Barlow, T. Kreouzis, and D. G.Lidzey, "High-speed electroluminescencemodulation of a conjugated-polymer lightemitting diode," Applied Physics Letters, vol.94, pp. 243301-3, 2009.

[20] P. A. Haigh, Z. Ghassemlooy, and I.Papakonstantinou, "1.4-Mb/s White OrganicLED Transmission System Using DiscreteMultitone Modulation," Photonics TechnologyLetters, IEEE, vol. 25, pp. 615-618, 2013.

[21] P. A. Haigh, Z. Ghassemlooy, I.Papakonstantinou, and M. Hoa Le, "2.7 Mb/sWith a 93-kHz White Organic Light EmittingDiode and Real Time ANN Equalizer,"Photonics Technology Letters, IEEE, vol. 25, pp.1687-1690, 2013.

[22] P. A. Haigh, F. Bausi, T. Kanesan, S. T.Le, S. Rajbhandari, Z. Ghassemlooy, et al.,"A 20-Mb/s VLC Link With a Polymer LED anda Multilayer Perceptron Equalizer,"Photonics Technology Letters, IEEE, vol. 26, pp.1975-1978, 2014.

[23] S. T. Le, T. Kanesan, F. Bausi, P. A.Haigh, S. Rajbhandari, Z. Ghassemlooy, etal., "10 Mb/s visible light transmissionsystem using a polymer light-emittingdiode with orthogonal frequency divisionmultiplexing," Optics Letters, vol. 39, pp.3876-3879, 2014/07/01 2014.

[24] C. J. Brabec, N. S. Sariciftci, and J. C.Hummelen, "Plastic Solar Cells," Adv. Funct.Mater., vol. 11, 2001.

[25] S. F. Tedde, J. Kern, T. Sterzl, J. Furst,P. Lugli, and O. Hayden, "Fully spraycoated organic photodiodes," Nano Lett, vol.9, pp. 980-3, Mar 2009.

[26] P. A. Haigh, Z. Ghassemlooy, L.-M. Hoa, S.Rajbhandari, F. Arca, S. F. Tedde, et al.,"Exploiting Equalization Techniques forImproving Data Rates in OrganicOptoelectronic Devices for Visible LightCommunications," Lightwave Technology, Journal of,vol. 30, pp. 3081-3088, 2012.

[27] Z. Ghassemlooy, P. A. Haigh, F. Arca, S.F. Tedde, O. Hayden, I. Papakonstantinou, etal., "Visible light communications: 3.75Mbits/s data rate with a 160 kHz bandwidthorganic photodetector and artificialneural network equalization [Invited],"Photonics Research, vol. 1, pp. 65-68,2013/08/01 2013.

[28] P. A. Haigh, Z. Ghassemlooy, I.Papakonstantinou, F. Arca, S. F. Tedde, O.Hayden, et al., "A 1-Mb/s Visible LightCommunications Link With Low BandwidthOrganic Components," Photonics TechnologyLetters, IEEE, vol. 26, pp. 1295-1298, 2014.

[29] A. H. Azhar, T. Tran, and D. O'Brien, "AGigabit/s Indoor Wireless TransmissionUsing MIMO-OFDM Visible-LightCommunications," Photonics Technology Letters,IEEE, vol. 25, pp. 171-174, 2013.

[30] V. S. C. Teichmann, A. N. Barreto, T. T.Pham, R. Rodes, I. T. Monroy, and D. A. A.Mello, "SC-FDE for MMF short reach opticalinterconnects using directly modulated 850nm VCSELs," Optics Express, vol. 20, pp.25369-25377, Nov 5 2012.

[31] H. Elgala and T. D. C. Little, "Polar-Based OFDM and SC-FDE Links Toward Energy-Efficient Gbps Transmission Under IM-DDOptical System Constraints [Invited],"Journal of Optical Communications and Networking,vol. 7, pp. A277-A284, 2015/02/01 2015.

[32] F. M. Wu, C. T. Lin, C. C. Wei, C. W.Chen, Z. Y. Chen, H. T. Huang, et al.,"Performance Comparison of OFDM Signal andCAP Signal Over High Capacity RGB-LED-Based WDM Visible Light Communication,"


Photonics Journal, IEEE, vol. 5, pp. 7901507-7901507, 2013.

[33] M. I. Olmedo, Z. Tianjian, J. B. Jensen,Z. Qiwen, X. Xiaogeng, S. Popov, et al.,"Multiband Carrierless Amplitude PhaseModulation for High Capacity Optical DataLinks," Lightwave Technology, Journal of, vol. 32,pp. 798-804, 2014.

[34] P. A. Haigh, S. T. Le, S. Zvanovec, Z.Ghassemlooy, P. Luo, T. Xu, et al., "Multi-band Carrier-less Amplitude and PhaseModulation for Bandlimited Visible LightCommunications Systems," IEEE Wireless Commun.Mag., in print,, 2015.

[35] P. A. Haigh, A. Burton, K. Werfli, H. L.Minh, E. Bentley, P. Chvojka, et al., "AMulti-CAP Visible Light CommunicationsSystem with 4.85 b/s/Hz SpectralEfficiency," IEEE J. Sel. Areas Commun., accepted,2015.

[36] IEEE 802.15 WPAN™ Task Group 7 (TG7) Visible LightCommunication. Available:http://www.ieee802.org/15/pub/TG7.html

[37] L. Kwonhyung, P. Hyuncheol, and J. R.Barry, "Indoor Channel Characteristics forVisible Light Communications,"Communications Letters, IEEE, vol. 15, pp. 217-219, 2011.

[38] J. R. Barry, J. M. Kahn, W. J. Krause, E.A. Lee, and D. G. Messerschmitt,"Simulation of multipath impulse responsefor indoor wireless optical channels,"Selected Areas in Communications, IEEE Journal on, vol.11, pp. 367-379, 1993.

[39] K.-D. Langer, J. Hilt, D. Shulz, F.Lassak, F. Hartlieb, C. Kottke, et al.,"Rate-adaptive visible light communicationat 500Mb/s arrives at plug and play,"Optoelectronics & Communications, SPIE Newsroom,2013.

[40] L. Grobe, A. Paraskevopoulos, J. Hilt, D.Schulz, F. Lassak, F. Hartlieb, et al.,"High-speed visible light communicationsystems," Communications Magazine, IEEE, vol.51, pp. 60-66, 2013.

[41] V. Jungnickel, V. Pohl, S. Nonnig, and C.von Helmolt, "A physical model of thewireless infrared communication channel,"Selected Areas in Communications, IEEE Journal on, vol.20, pp. 631-640, 2002.

[42] S. Jivkova and M. Kavehrad, "Shadowing andblockage in indoor optical wirelesscommunications," in Global TelecommunicationsConference, 2003. GLOBECOM '03. IEEE, 2003, pp.3269-3273 vol.6.

[43] T. Komine and M. Nakagawa, "A study ofshadowing on indoor visible-light wireless

communication utilizing plural white LEDlightings," in Wireless Communication Systems,2004, 1st International Symposium on, 2004, pp.36-40.

[44] P. Chvojka, S. Zvanovec, P. A. Haigh, andZ. Ghassemlooy, "Channel Characteristicsof Visible Light Communications withinDynamic Indoor Environment," LightwaveTechnology, Journal of, vol. PP, pp. 1-1, 2015.

[45] L. Hugh Sing and G. Pang, "Positioningbeacon system using digital camera andLEDs," Vehicular Technology, IEEE Transactions on,vol. 52, pp. 406-419, 2003.

[46] Katayama; Mutsumi, M. Kazuyuki, and K.Kazumitsu, "Vehicle position detectionsystem " JP Patent, 2010.

[47] A. Arafa, J. Xian, and R. Klukas,"Wireless Indoor Optical Positioning Witha Differential Photosensor," PhotonicsTechnology Letters, IEEE, vol. 24, pp. 1027-1029,2012.

[48] L. Pengfei, Z. Ghassemlooy, M. Hoa Le, A.Khalighi, Z. Xiang, Z. Min, et al.,"Experimental demonstration of an indoorvisible light communication positioningsystem using dual-tone multi-frequencytechnique," in Optical Wireless Communications(IWOW), 2014 3rd International Workshop in, 2014,pp. 55-59.

[49] T. Q. Wang, Y. A. Sekercioglu, A. Neild,and J. Armstrong, "Position Accuracy ofTime-of-Arrival Based Ranging UsingVisible Light With Application in IndoorLocalization Systems," Lightwave Technology,Journal of, vol. 31, pp. 3302-3308, 2013.

[50] J. B. Carruthers, S. M. Caroll, and P.Kannan, "Propagation modelling for indooroptical wireless communications using fastmulti-receiver channel estimation," IEEProceedings- Optoelectronics, vol. 150, pp. 473-481, 2003.

[51] T. S. Rappaport, Wireless Communications:Prentice-Hall, 2002.

[52] R. N. McDonough and A. D. Whalen, Detectionof Signals in Noise. San Diego, CA, USA: Wiley,1995.

[53] D. Wu, Z. Ghassemlooy, W.-d. Zhong, M.-A.Khalighi, H. L. Minh, C. Chen, et al.,"Effect of Optimal Lambertian Order on thePerformance of Cellular Indoor OpticalWireless Communications and Positoning "Journal of Lightwave Technologies, 2015,submitted...

[54] (2014, 10. Feb). The LED Streetlight ReplacementProgram Available:http://bsl.lacity.org/led.html

http://bsl.lacity.org/led.html

http://www.ieee802.org/15/pub/TG7.html


[55] P. Luo, Z. Ghassemlooy, H. L. Minh, E.Bentley, A. Burton, and X. Tang,"Performance analysis of a car-to-carvisible light communication system,"Applied Optics, vol. 54, pp. 1696-1706,2015/03/01 2015.

[56] I. Takai, S. Ito, K. Yasutomi, K. Kagawa,M. Andoh, and S. Kawahito, "LED and CMOSImage Sensor Based Optical WirelessCommunication System for AutomotiveApplications," Photonics Journal, IEEE, vol. 5,pp. 6801418-6801418, 2013.

[57] T. Nagura, T. Yamazato, M. Katayama, T.Yendo, T. Fujii, and H. Okada, "Trackingan LED array transmitter for visible lightcommunications in the driving situation,"in Wireless Communication Systems (ISWCS), 2010 7thInternational Symposium on, 2010, pp. 765-769.

About Authors ...Stanislav ZVANOVEC received M.Sc. in2002, a Ph.D. in 2006 at the CzechTechnical University in Prague. Now heis a full professor and vice-head ofthe Department of Electromagnetic Fieldat the Faculty of ElectricalEngineering, Czech Technical Universityin Prague. He leads a Free-space andfiber optics team from Faculty ofElectrical Engineering, CTU and severalresearch projects. His current researchinterests include wireless opticalcommunications, visible lightcommunications, remote sensing andoptical fiber sensors.Petr CHVOJKA was born in 1987. Hereceived M.Sc. in 2013 at the CzechTechnical University in Prague. Now heis a postgraduate student and aresearcher at the Department ofElectromagnetic Fields, Czech TechnicalUniversity in Prague, where he is amember of a Free-space and Fiber Opticsteam. His research area includesvisible light communications, OLEDtechnologies and wireless opticalcommunications.Paul Anthony HAIGH received the PhDdegree in visible light communications

from Northumbria University in 2014,publishing 13 articles in high rankingjournals. Between 2010 and 2011 he heldthe prestigious Marie Curie Fellowshipat CERN where he worked onoptoelectronic links for large hadroncollider experiments. In 2010 Paulreceived the BEng (Hons) degree inCommunications Engineering fromNorthumbria University. Currently, heis a Research Associate within the HighPerformance Networks group at theUniversity of Bristol working on theEPSRC TOUCAN project. His research isfocused on real time seamless,transparent and adaptive andprogrammable interfaces betweenwireless and wired multi-technologynetworks.Zabih GHASSEMLOOY, CEng, Fellow of IET,Senior Member of IEEE received his BSc(Hons) from the Manchester MetropolitanUniversity in 1981, and MSc and PhDfrom the University of ManchesterInstitute of Science and Technology(UMIST), in 1984 and 1987,respectively. 1986-87 worked in UMISTand from 1987 to 1988 was a Post-doctoral Research Fellow at the CityUniversity, London. 1988 joinedSheffield Hallam University as aLecturer, becoming a Professor inOptical Communications in 1997. 2004-2012 was an Associate Dean for Researchin the School of Computing, Engineeringand from 2012-2014 Associate Dean forResearch and Innovation in the Facultyof Engineering and Environment,Northumbria University at Newcastle,UK. He currently heads the NorthumbriaCommunications Research Laboratorieswithin the Faculty. He is the Editor-in-Chief of the International Journalof Optics and Applications, and BritishJournal of Applied Science &Technology. His researches interestsare on optical wireless communications,visible light communications and radioover fibre/free space optics. He hassupervised over 48 PhD students and


published more than 550 papers (195 injournals + 4 books). He is a co-authorof a CRC book on “Optical WirelessCommunications – Systems and ChannelModelling with Matlab (2012). From2004-06 he was the IEEE UK/IR

Communications Chapter Secretary, theVice-Chairman (2004-2008), the Chairman(2008-2011), and Chairman of the IETNorthumbria Network (Oct 2011-).

Visible Light Communications towards 5G

Documents

Transcript of Visible Light Communications towards 5G