Full-Text PDF - downloads.hindawi.comdownloads.hindawi.com/journals/wcmc/2017/4094096.pdf ·...

Research ArticleSecrecy Dimming Capacity in Multi-LED PAM-BasedVisible Light Communications

Byung Wook Kim

Department of ICT Automotive Engineering Hoseo University 201 Sandan 7-ro Seongmun-myeon Dangjin-siChungcheongnam-do 31702 Republic of Korea

Correspondence should be addressed to Byung Wook Kim philip0110gmailcom

Received 8 June 2017 Accepted 31 July 2017 Published 28 August 2017

Academic Editor Shingo Yamaguchi

Copyright copy 2017 Byung Wook Kim This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Recently mobile cloud computing (MCC) has gained a lot of interest for researchers building the next-generation mobileapplications Because unauthorized access may cause serious problems security and privacy with MCC have become significantissuesThis paper addresses the secrecy dimming capacity of secure transmission in MCC over visible light communication (VLC)channels By obtaining the entropy-maximizing symbol probability of multiple light emitting diode- (LED-) based pulse amplitudemodulation (PAM) mathematical analysis of the secrecy dimming capacity of VLC was derived Simulation results show that thesecure transmission ability of multi-LED-based VLC is determined according to the number of activated LEDs and target dimminglevel This can be a guideline for practical VLC-based mobile network designers intending to secure wireless transmission and todecide on the number of activated LEDs at target dimming level to operate

1 Introduction

With the proliferation of smart mobile devices and cloudcomputing technologies mobile cloud computing (MCC)has emerged as one of the most important technologies fornext-generation mobile services [1ndash4] MCC is a technol-ogy that combines mobile devices and cloud computing toperform both data storage and data processing outside themobile device The use of MCC can offer ubiquitous on-demand network access to a shared pool of configurabledata processing resources that can be rapidly provided withminimal management effort on the user side and for theservice provider

As the radio frequency spectrum of mobile devicesbecomes increasingly crowded and light emitting diodes(LEDs) are more widely used for various incident lightapplications visible light communication (VLC) is gainingappeal as an alternative to conventional radio frequency (RF)for billions ofmobile devices that need to be networked [5ndash9]A VLC system can transmit information by modulating theintensity of LEDs at high frequencies (ge150Hz) so that LEDflickering is unnoticeable to the human eye Due to VLCrsquoslarge spectral availability and the free unlicensed spectrum

mobile phone-based VLC can be an important candidatefor MCC applications By adapting multiple-input multiple-output (MIMO) technology in VLC where multiple LEDsandmultiple photo-detectors (PDs) are utilized independentdata streams are simultaneously transmitted from all lightsources and thus it promises a further increase in bandwidthfor MCC networks

Although the widespread use of mobile devices cancontribute to the improvement of standard of living anxietyabout the leakage of personal information is increasingWhile the large spectral availability of the visible lightspectrum is certainly themain reason for the growing interestin VLC the inherent security that stems from line-of-sightpropagation and the nonpenetrating nature of light waves isalso an important issue and the most captivating differencecompared to RF This vulnerability may be even worse forRF communications due to the nature of the broadcast ofthe radio propagation and the inherent random nature ofthe radio channel Although VLC channel exhibits perfectsecurity in a private-room security of the transmitted signalin public areas such as hotels stations libraries or planescannot be guaranteed and thus it makes transmissions overVLC channel vulnerable to unexpected wiretappersrsquo attacks

HindawiWireless Communications and Mobile ComputingVolume 2017 Article ID 4094096 6 pageshttpsdoiorg10115520174094096

2 Wireless Communications and Mobile Computing

Because userrsquos privacy and integrity of data are importantsecure communication for VLC becomes a significant topicfor both academia and industry

Recently there have been several studies that addresssecurity in VLC systems Mostafa and Lampe addressedsecure VLC link at the physical layer [10] by investigating theachievable secrecy rates of aGaussianwiretap channel Cho etal [11] investigated secrecy outage probability of the downlinkfor VLC Zhang et al [12] proposed a secure system forbarcode-based VLC that is for secure transmission betweena screen and a camera To support secure data exchangehowever the system requires a fully duplex VLC channelTo the best of our knowledge the secrecy dimming capacityof VLC using multiple LEDs has not been studied in theliterature

This paper investigates the secrecy dimming capacityof pulse amplitude modulation- (PAM-) based VLC usingmultiple LEDs To derive a secrecy dimming capacity theentropy-maximizing PAM symbol probabilities were math-ematically derived Then the secrecy dimming capacity formultiple LED-based VLC systems was analyzed Simulationresults demonstrated the baseline information of the secrecydimming capacity of multi-LED PAM-based VLC systemsTo obtain robust secure transmission underVLC the numberof activated LEDs should be decided according to the targetdimming level

The main contribution of this paper is the analyticalderivation of secrecy dimming capacity in the presence ofthe wiretapper The curves of secrecy dimming capacity wereplotted against the SNR and target dimming ratio of themain channel in the presence of the wiretappers channelThiscan be a guideline for practical VLC-based mobile networkdesigners intending to secure wireless transmission and todecide on the number of activated LEDs at target dimminglevel to operate This paper paves the way for a new studyon the secrecy dimming capacity when both the main andwiretapper VLC channels are considered

The remainder of this paper is organized as follows Thesystem model of a multi-LED PAM-based VLC scheme ispresented in Section 2 Section 3 addresses derivation ofentropy-maximizing PAM symbol probabilities Section 4reports analysis of secrecy dimming capacity and evaluationSection 5 presents the conclusions

2 System Model of Multi-LEDPAM-Based VLC

A VLC scenario with one transmitter and one legitimatereceiver in the presence of a wiretapper where the solidand dash lines represent the main channel (from transmit-ter to legitimate receiver) and the wiretap channel (fromtransmitter to wiretapper) respectively is shown in Figure 1When a transmitter sends its signal to a legitimate receivera wiretapper may overhear such transmission due to thenature of wireless medium Considering the fact that recentmobile devices are highly standardized wiretappers caneasily acquire communication parameters such as signalwaveform coding modulation scheme and encryption algo-rithm In addition the secret key can be accessed by the

wiretapper through the exhaustive search In this scenariothe transmitted signal can be interpreted at the wiretapperby decoding its overheard signal leading the legitimatetransmission to be insecure

The number of activated LEDs 119873119905 at the transmitter isassumed to be same as the number of PDs at legitimate andwiretap receivers When the sender transmits119873119905 times 1 symbolvector X the119873119905 times 1 received symbol vectors Y1 and Y2 at thelegitimate receiver and the wiretapper respectively are givenby

Y1 = X + Z1Y2 = X + Z2 (1)

Z119902 119902 = 1 2 is119873119905 times 1 additive white Gaussian noise (AWGN)vector following Z119902 sim 119873(0K119902) where

K119902 = [[[[[

1205902119902 sdot sdot sdot 0 d

0 sdot sdot sdot 1205902119902

]]]]] (2)

which satisfies

0 lt K1 lt K2 (3)

Because the goal of the transmitter here is not just to conveyinformation reliably to the legitimate receiver but also toperfectly secure the data from the wiretap receiver thecondition in (3) should be satisfied

Note that intensity modulation with direct detection(IMDD) is typically used in VLC systems where signals aretransmitted through an LED in the form of optical power Inthe receiver a PD is employed to convert the optical powersignal into electrical signals By applying MIMO technologyto VLC independent data streams are simultaneously emit-ted from multiple LEDs

The PAM modulation method is considered for datamodulation where the information is encoded in the ampli-tude of a series of signal pulses Applying MIMO conceptto VLC the independent PAM-modulated signals are trans-mitted from activated LEDs As this method simultaneouslyemits the identical signal from several activated LEDsthe optical transmission power is assumed to be equallydistributed across all LEDs For the individual LED theintensities of PAM with119872th order modulation are given by

119868119899 = 2119868119899119872 + 1 119899 = 1 2 119872 (4)

where 119868 is medium intensity of PAM signal candidatesSignals with intensity 119868119899 = 0 at an individual LED are not usedfor the signal modulation so that activated and nonactivatedLEDs can be distinguished

3 Entropy-Maximizing Symbol Probability ofMulti-LED PAM

To use LEDs in a mobile device as an illumination source thetarget dimming level should be determined according to var-ious illuminating applications To meet the target dimming

Wireless Communications and Mobile Computing 3

Data Data

Data

Modulator Demodulator

Demodulator

LEDdriver

Transmitter

+

+

PD

PD

Wiretapper

Legitimate receiverY1

Z1

Y2

Z2

X

Figure 1 The system model of VLC in the presence of wiretapper

level ofmultiple LED-basedVLC systems symbol probabilityof various PAM-levels should be different Because there aremany possible alternatives for meeting the given target dim-ming level each will result in a different entropy Thereforean entropy-maximizing probability distribution is derived todetermine secrecy dimming capacity The entropy of PAM atan individual LED is given by

minus119872sum119899=1

119901 (119899) log2119901 (119899) (5)

where 119872 is the order of modulation and 119901(119899) is the 119899thsymbol probability

To meet the total transmit power constraint that is thedimming level constraint each respective emitted intensityis the total power divided by the number of activated LEDsBy doing this the total optical power emitted is constantregardless of the number of activated LEDs Using thenormalization constraint associated with the definition ofa probability density the intensity of an individual LEDconsidering dimming constraint is written as

119872sum119899=1

2119868119899119872 + 1119901 (119899) = 119863119905119873119905 2119868 (6)

where 119863119905 is the target dimming ratio and 119873119905 is the numberof activated LEDs Note that the range of the target dimmingratio is 0 le 119863119905 le 1 and target dimming level is set to119863119905times100Due to the sum-power constraint on the transmitting LEDsthe intensity is divided by 119873119905 Using the symbol probabilityset (119901(1) 119901(2) 119901(119899)) satisfying (6) we obtain symbolprobability that maximizes (5) by using Lagrange multipliersNote that objective function (5) is concave and the constraintfunction (6) is linear Therefore the Lagrange equation ispresented as

119871 (119901 (1) 119901 (2) 119901 (119872) 120573 120574)= minus119872sum119899=1

119901 (119899) log2119901 (119899) minus 120573(119872sum119899=1

119901 (119899) minus 1)

minus 120574(119872sum119899=1

2119868119899119872 + 1119901 (119899) minus 2119868119863119905119873119905) (7)

where 120573 and 120574 are the scalar Lagrange coefficients The prob-lem of maximizing (7) is solved by converting the problemto an optimization problem with independent variables The

variables to this problem are 119901(119899) 120573 120574 and three equationsare obtained from the gradient with respect to these variablesHaving the same number of equations and variables makesthe problem determined and can be solved To calculate thegradient of the Lagrange equation the functional derivativeon (7) with respect to the variables [13] 119901(119899) 120573 120574 is taken asfollows

120597119871120597119901 (119899) = minuslog2119901 (119899) minus 1ln 2 minus 120573 minus 120574 2119868119899119873119905 (119872 + 1) = 0 (8)

120597119871120597120573 = 119872sum119899=1

119901 (119899) minus 1 = 0 (9)

120597119871120597120574 = 119872sum119899=1

2119868119899119872 + 1119901 (119899) minus 2119868119863119905119873119905 = 0 (10)

Equation (8) is rearranged as

119901 (119899) = 2minus1 ln 2minus120573minus120574(2119868119899(119872+1)) (11)

Equations (9) and (11) give

2119886 = 119903 (1 minus 119903119872)1 minus 119903 (12)

where

119886 = 1ln 2 + 120573

119903 = 2minus1205742119868(119872+1)(13)

Equations (10) and (11) are simplified as

119863119905= 1198731199052minus119886(119872 + 1) (

119903 (1 minus 119903119872minus1)(1 minus 119903)2 minus (119872 minus 1) 119903119872(1 minus 119903) + 119872119903119872) (14)

From (12) and (13) multiple solutions for (120573 120574) are obtainedThen 120573 and 120574 solutions are carefully chosen to make thegroup of symbol probabilities (119901(1) 119901(2) 119901(119899)) non-negative and real This yields (11) providing PAM symbolprobabilities thatmaximize entropywhile satisfying the giventarget dimming level

Figure 2 presents PAM symbol probabilities maximizingentropy when 4-PAM is considered To meet the constraintof target dimming ratio turn-on probability maximizingentropy for each PAM signal level is different except whentarget dimming ratio is 05


0

01

02

03

04

05

06

07

08

09

1

Sym

bol p

roba

bilit

y

03 04 05 06 07 0802Dimming ratio

p(1)

p(2)

p(3)

p(4)

Figure 2 Entropy-maximizing symbol probability (4-PAM)

4 Analysis of Secrecy Dimming Capacity forMulti-LED PAM-Based VLC

In this section the secrecy dimming capacity for multi-LED PAM-based VLC is analyzed The dimming capacitymeans the achievable data rate obtained by a specific mod-ulation and dimming condition and expressed as mutualinformation The secrecy dimming capacity is the differencein the dimming capacities of the main channel and thewiretapperrsquos channel under the constraint of target dimminglevel [14] When the main channel is less noisy comparedto the wiretapperrsquos channel the secrecy dimming capacity isgiven by

119862119889 = max119901(119899)

[119868 (XY1) minus 119868 (XY2)] (15)

Note that the secrecy dimming capacity can be expressedin even simpler terms in certain cases When 119868(XY1) and119868(XY2) can be individually maximized by the same 119901(119899)the secrecy capacity is simply the difference in channelcapacities [15] As the mutual information between tworandom variables is a measure of the amount of informationthey contain about each other the difference 119868(XY1) minus119868(XY2) represents the extra information that Y1 shares withX over that which X and Y2 share Because this approachspecializes the setting to less noisy main channel by imposingthe restriction of K2 gt K1 secrecy dimming capacity isalways positive If the secrecy capacity falls below zero thetransmission from transmitter to legitimate receiver becomesinsecure and the wiretapper would succeed in interceptingthe transmitted information In order to enhance transmis-sion security against a wiretapperrsquos attack it is of importanceto reduce the probability of occurrence of an intercept eventthrough enlarging the secrecy dimming capacity

The expression of secrecy dimming capacity becomes justthe difference in the mutual information of the main andthe wiretapperrsquos channel Because the mutual information is

defined as differential entropy the secrecy dimming capacityis presented as

119862119889 = 119868 (XY1) minus 119868 (XY2)= ℎ (Y1) minus ℎ (Y1 | X) minus ℎ (Y2) minus ℎ (Y2 | X)= ℎ (Y1) minus ℎ (Z1) minus ℎ (Y2) minus ℎ (Z2)= minusintinfinminusinfin

119875119884 (Y1) log2119875119884 (Y1) 119889Y1minus 12 log2 (2120587119890)119873119905 1003816100381610038161003816K11003816100381610038161003816minus intinfinminusinfin

119875119884 (Y2) log2119875119884 (Y2) 119889Y2minus 12 log2 (2120587119890)119873119905 1003816100381610038161003816K21003816100381610038161003816

(16)

The probability distribution functions of the transmittedsignal 119875119883(X) the received signal via main channel 119875119884(Y1)the received signal via wiretap channel 119875119884(Y2) AWGNvia main channel 119875119885(Z1) and AWGN via wiretap channel119875119885(Z2) are expressed as

119875119883 (X) = 119872sum119899=1

119901 (119899) 120575 (X minus b119899)

119875119884 (Y1) = 119872sum119899=1

119901 (119899) 119875119885 (Y1 minus b119899)

119875119884 (Y2) = 119872sum119899=1

119901 (119899) 119875119885 (Y2 minus b119899) 119875119885 (Z1) = 1120587119873119905 det (C119885) exp (minusZ1

119879Cminus1119885 Z1) 119875119885 (Z2) = 1120587119873119905 det (C119885) exp (minusZ2

119879Cminus1119885 Z2)

(17)

where b119899 is the 119873119905 times 1 intensity vector with elementsof 119868119899 From (16) the secrecy dimming capacity with themaximum information rate of themain channel (transmitter-to-legitimate receiver) with the ignorance at the wiretappercan be obtained

Figure 3 shows the secrecy dimming capacity of 4-PAMversus SNR values with different target dimming level Theratio of average legitimate noise variance to wiretappersaverage noise variance is defined as 120572 = 1205902112059022 We can seethat secrecy dimming capacity for 120572 = 04 is larger than thatof 120572 = 06 This means that the more the noise on wiretapchannel is the greater the secrecy dimming capacity can beachieved When dimming level is 44 a single activatedLED shows a robust secrecy dimming capacity compared tomultiple activated LEDs For 80 dimming level scenariohowever multiple activated LEDs that is 119873119905 = 2 3 resultin better performance of secrecy dimming capacity than thescenario with single activated LED

Figure 4 shows the secrecy dimming capacity of 4-PAM versus target dimming ratio For 120572 = 06 the single


Secrecy dimming capacity (4-PAM 44 dimming)

minus5 0 5 10 15 20 25minus10SNR (dB)

0

01

02

03

04

05

06

07Se

crec

y di

mm

ing

capa

city

(bits

sym

bol)

Nt = 1 ( = 04)Nt = 2 ( = 04)Nt = 3 ( = 04)

Nt = 1 ( = 06)Nt = 2 ( = 06)Nt = 3 ( = 06)

(a) Target dimming level = 44


0

02

04

06

08

1

12

14

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)


Nt = 1 ( = 04)Nt = 2 ( = 04)Nt = 3 ( = 04)

Nt = 1 ( = 06)Nt = 2 ( = 06)Nt = 3 ( = 06)

(b) Target dimming level = 80

Figure 3 Secrecy dimming capacity with 44 and 80 of target dimming level

Secrecy dimming capacity (4-PAM SN２ = 10 dB)

0

005

01

015

02

025

03

035

04

045

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)

045 05 055 06 065 07 075 0804Dimming ratio

Nt = 1

Nt = 2

Nt = 3

(a) 120572 = 04


Nt = 1

Nt = 2

Nt = 3

0

005

01

015

02

025

03

035

04

045

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)

045 05 055 06 065 07 075 0804Dimming ratio

(b) 120572 = 06

Figure 4 Secrecy dimming capacity with 120572 = 04 and 120572 = 06

activated LED shows best performance when target dimmingratio is le048 When target dimming ratio is larger than048 the performance with dual activated LEDs outperformsother scenarios As the target dimming ratio approaches08 the use of two and three activated LEDs shows similarsecrecy dimming capacity performance For 120572 = 04 thesingle activated LED shows best performance when targetdimming ratio is le058 If the target dimming ratio is set

to a value larger than 058 best secrecy dimming capacityperformance can be achieved when dual activated LEDsare considered From Figures 4(a) and 4(b) we see thatthe use of multiple activated LEDs results in robust secrecydimming capacity performance when target dimming ratiois set to high value When low dimming is required the useof a single activated LED guarantees best secrecy dimmingcapacity


5 Conclusions

Because of the increased demand for processing and storagecapabilities for mobile devices MCC is gaining popularityAs MCC makes data storage and data processing possibleoutside of a mobile device security risk becomes a significantissue This paper investigates the secrecy dimming capacityof PAM in VLC with multiple LED arrays By obtainingentropy-maximizing symbol probabilities based on multipleLED arrays the secrecy dimming capacity considering thepresence of wiretapper is mathematically derived Simulationresults showed the baseline information of secure capacityperformance of PAM-based VLC systems According to thetarget dimming ratio the number of activated LEDs shouldbe decided to guarantee robust secrecy dimming capacityThe result can be a guideline to practical VLC-based mobilesystems intending to secure wireless transmission

Most of the existing works of secure transmission ofVLC have neglected the joint consideration of various formsof the wireless attacks including both eavesdropping anddenial-of-service (DoS) behaviors It will be very importantto explore new techniques of joint defense and maximizingsecrecy dimming capacity against multiple different wirelessattacks Furthermore it will be important to consider jointoptimization of security and throughput of the VLC systemwhich is a problem to be solved in the future

Conflicts of Interest

The author declares that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

This work was supported by the National Research Founda-tion of Korea (NRF) grant funded by the Korean government(MSIP) (2016R1C1B1013942)

References

[1] Z Sanaei S Abolfazli A Gani and R Buyya ldquoHeterogeneity inmobile cloud computing taxonomy and open challengesrdquo IEEECommunications Surveys ampTutorials vol 16 no 1 pp 369ndash 3922014

[2] H T Dinh C Lee D Niyato and P Wang ldquoA survey of mobilecloud computing Architecture applications and approachesrdquoWireless Communications andMobile Computing vol 13 no 18pp 1587ndash1611 2013

[3] J Bou Abdo T Bourgeau J Demerjian and H ChaouchildquoExtended privacy in crowdsourced location-based servicesusing mobile cloud computingrdquo Mobile Information Systemsvol 2016 Article ID 7867206 13 pages 2016

[4] M Alizadeh S Abolfazli M Zamani S Baaaharun and KSakurai ldquoAuthentication in mobile cloud computing A surveyrdquoJournal of Network and Computer Applications vol 61 pp 59ndash80 2016

[5] D Karunatilaka F Zafar V Kalavally and R Parthiban ldquoLEDbased indoor visible light communications state of the artrdquoIEEE Communications Surveys and Tutorials vol 17 no 3 pp1649ndash1678 2015

[6] A Nuwanpriya S-W Ho and C S Chen ldquoIndoor MIMOVisi-ble Light Communications Novel Angle Diversity Receivers forMobile Usersrdquo IEEE Journal on Selected Areas in Communica-tions vol 33 no 9 pp 1780ndash1792 2015

[7] P H Pathak X Feng P Hu and P Mohapatra ldquoVisible lightcommunication networking and sensing a survey potentialand challengesrdquo IEEE Communications Surveys amp Tutorials vol17 no 4 pp 2047ndash2077 2015

[8] L Yin W O Popoola X Wu and H Haas ldquoPerformanceevaluation of non-orthogonal multiple access in visible lightcommunicationrdquo IEEE Transactions on Communications vol64 no 12 2016

[9] A C Boucouvalas P Chatzimisios Z Ghassemlooy M Uysaland K Yiannopoulos ldquoStandards for indoor Optical WirelessCommunicationsrdquo IEEECommunicationsMagazine vol 53 no3 pp 24ndash31 2015

[10] A Mostafa and L Lampe ldquoPhysical-layer security for indoorvisible light communicationsrdquo in Proceedings of the 1st IEEEInternational Conference on Communications (ICC rsquo14) pp3342ndash3347 Sydney Australia June 2014

[11] S Cho G Chen and J P Coon ldquoSecrecy analysis in visible lightcommunication systemswith randomly located eavesdroppersrdquoin Proceedings of the 2017 IEEE International Conference onCommunications Workshops (ICC Workshops) pp 475ndash480Paris France May 2017

[12] B Zhang K Ren G Xing X Fu and CWang ldquoSBVLC Securebarcode-based visible light communication for smartphonesrdquoIEEE Transactions on Mobile Computing vol 15 no 2 pp 432ndash446 2016

[13] D P Bertsekas Constrained Optimization and Lagrange Multi-plier Methods Academic Press New York NY USA 1982

[14] Y Zou J Zhu X Wang and V C M Leung ldquoImprovingphysical-layer security in wireless communications using diver-sity techniquesrdquo IEEE Network vol 29 no 1 pp 42ndash48 2015

[15] M VanDijk ldquoOn a special class of broadcast channels with con-fidential messagesrdquo IEEE Transactions on Information Theoryvol 43 no 2 pp 712ndash714 1997

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of


International Journal of

RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal of

Volume 201

Submit your manuscripts athttpswwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



Because userrsquos privacy and integrity of data are importantsecure communication for VLC becomes a significant topicfor both academia and industry

Recently there have been several studies that addresssecurity in VLC systems Mostafa and Lampe addressedsecure VLC link at the physical layer [10] by investigating theachievable secrecy rates of aGaussianwiretap channel Cho etal [11] investigated secrecy outage probability of the downlinkfor VLC Zhang et al [12] proposed a secure system forbarcode-based VLC that is for secure transmission betweena screen and a camera To support secure data exchangehowever the system requires a fully duplex VLC channelTo the best of our knowledge the secrecy dimming capacityof VLC using multiple LEDs has not been studied in theliterature

This paper investigates the secrecy dimming capacityof pulse amplitude modulation- (PAM-) based VLC usingmultiple LEDs To derive a secrecy dimming capacity theentropy-maximizing PAM symbol probabilities were math-ematically derived Then the secrecy dimming capacity formultiple LED-based VLC systems was analyzed Simulationresults demonstrated the baseline information of the secrecydimming capacity of multi-LED PAM-based VLC systemsTo obtain robust secure transmission underVLC the numberof activated LEDs should be decided according to the targetdimming level

The main contribution of this paper is the analyticalderivation of secrecy dimming capacity in the presence ofthe wiretapper The curves of secrecy dimming capacity wereplotted against the SNR and target dimming ratio of themain channel in the presence of the wiretappers channelThiscan be a guideline for practical VLC-based mobile networkdesigners intending to secure wireless transmission and todecide on the number of activated LEDs at target dimminglevel to operate This paper paves the way for a new studyon the secrecy dimming capacity when both the main andwiretapper VLC channels are considered

The remainder of this paper is organized as follows Thesystem model of a multi-LED PAM-based VLC scheme ispresented in Section 2 Section 3 addresses derivation ofentropy-maximizing PAM symbol probabilities Section 4reports analysis of secrecy dimming capacity and evaluationSection 5 presents the conclusions

2 System Model of Multi-LEDPAM-Based VLC

A VLC scenario with one transmitter and one legitimatereceiver in the presence of a wiretapper where the solidand dash lines represent the main channel (from transmit-ter to legitimate receiver) and the wiretap channel (fromtransmitter to wiretapper) respectively is shown in Figure 1When a transmitter sends its signal to a legitimate receivera wiretapper may overhear such transmission due to thenature of wireless medium Considering the fact that recentmobile devices are highly standardized wiretappers caneasily acquire communication parameters such as signalwaveform coding modulation scheme and encryption algo-rithm In addition the secret key can be accessed by the

wiretapper through the exhaustive search In this scenariothe transmitted signal can be interpreted at the wiretapperby decoding its overheard signal leading the legitimatetransmission to be insecure

The number of activated LEDs 119873119905 at the transmitter isassumed to be same as the number of PDs at legitimate andwiretap receivers When the sender transmits119873119905 times 1 symbolvector X the119873119905 times 1 received symbol vectors Y1 and Y2 at thelegitimate receiver and the wiretapper respectively are givenby

Y1 = X + Z1Y2 = X + Z2 (1)

Z119902 119902 = 1 2 is119873119905 times 1 additive white Gaussian noise (AWGN)vector following Z119902 sim 119873(0K119902) where

K119902 = [[[[[

1205902119902 sdot sdot sdot 0 d

0 sdot sdot sdot 1205902119902

]]]]] (2)

which satisfies

0 lt K1 lt K2 (3)

Because the goal of the transmitter here is not just to conveyinformation reliably to the legitimate receiver but also toperfectly secure the data from the wiretap receiver thecondition in (3) should be satisfied

Note that intensity modulation with direct detection(IMDD) is typically used in VLC systems where signals aretransmitted through an LED in the form of optical power Inthe receiver a PD is employed to convert the optical powersignal into electrical signals By applying MIMO technologyto VLC independent data streams are simultaneously emit-ted from multiple LEDs

The PAM modulation method is considered for datamodulation where the information is encoded in the ampli-tude of a series of signal pulses Applying MIMO conceptto VLC the independent PAM-modulated signals are trans-mitted from activated LEDs As this method simultaneouslyemits the identical signal from several activated LEDsthe optical transmission power is assumed to be equallydistributed across all LEDs For the individual LED theintensities of PAM with119872th order modulation are given by

119868119899 = 2119868119899119872 + 1 119899 = 1 2 119872 (4)

where 119868 is medium intensity of PAM signal candidatesSignals with intensity 119868119899 = 0 at an individual LED are not usedfor the signal modulation so that activated and nonactivatedLEDs can be distinguished

3 Entropy-Maximizing Symbol Probability ofMulti-LED PAM

To use LEDs in a mobile device as an illumination source thetarget dimming level should be determined according to var-ious illuminating applications To meet the target dimming


Data Data

Data


Demodulator

LEDdriver

Transmitter

+

+

PD

PD

Wiretapper


Z1

Y2

Z2

X



minus119872sum119899=1

119901 (119899) log2119901 (119899) (5)



119872sum119899=1

2119868119899119872 + 1119901 (119899) = 119863119905119873119905 2119868 (6)


119871 (119901 (1) 119901 (2) 119901 (119872) 120573 120574)= minus119872sum119899=1

119901 (119899) log2119901 (119899) minus 120573(119872sum119899=1

119901 (119899) minus 1)

minus 120574(119872sum119899=1

2119868119899119872 + 1119901 (119899) minus 2119868119863119905119873119905) (7)




120597119871120597120573 = 119872sum119899=1

119901 (119899) minus 1 = 0 (9)

120597119871120597120574 = 119872sum119899=1

2119868119899119872 + 1119901 (119899) minus 2119868119863119905119873119905 = 0 (10)




2119886 = 119903 (1 minus 119903119872)1 minus 119903 (12)

where

119886 = 1ln 2 + 120573

119903 = 2minus1205742119868(119872+1)(13)


119863119905= 1198731199052minus119886(119872 + 1) (





0

01

02

03

04

05

06

07

08

09

1

Sym

bol p

roba

bilit

y

03 04 05 06 07 0802Dimming ratio

p(1)

p(2)

p(3)

p(4)




119862119889 = max119901(119899)

[119868 (XY1) minus 119868 (XY2)] (15)







(16)


119875119883 (X) = 119872sum119899=1

119901 (119899) 120575 (X minus b119899)

119875119884 (Y1) = 119872sum119899=1

119901 (119899) 119875119885 (Y1 minus b119899)

119875119884 (Y2) = 119872sum119899=1



119879Cminus1119885 Z2)

(17)







0

01

02

03

04

05

06

07Se

crec

y di

mm

ing

capa

city

(bits

sym

bol)

Nt = 1 ( = 04)Nt = 2 ( = 04)Nt = 3 ( = 04)

Nt = 1 ( = 06)Nt = 2 ( = 06)Nt = 3 ( = 06)



0

02

04

06

08

1

12

14

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)


Nt = 1 ( = 04)Nt = 2 ( = 04)Nt = 3 ( = 04)

Nt = 1 ( = 06)Nt = 2 ( = 06)Nt = 3 ( = 06)




0

005

01

015

02

025

03

035

04

045

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)

045 05 055 06 065 07 075 0804Dimming ratio

Nt = 1

Nt = 2

Nt = 3

(a) 120572 = 04


Nt = 1

Nt = 2

Nt = 3

0

005

01

015

02

025

03

035

04

045

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)

045 05 055 06 065 07 075 0804Dimming ratio

(b) 120572 = 06





5 Conclusions





Acknowledgments


References
















RoboticsJournal of





Journal of



RotatingMachinery



Journal of

Volume 201


VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Data Data

Data


Demodulator

LEDdriver

Transmitter

+

+

PD

PD

Wiretapper


Z1

Y2

Z2

X



minus119872sum119899=1

119901 (119899) log2119901 (119899) (5)



119872sum119899=1

2119868119899119872 + 1119901 (119899) = 119863119905119873119905 2119868 (6)


119871 (119901 (1) 119901 (2) 119901 (119872) 120573 120574)= minus119872sum119899=1

119901 (119899) log2119901 (119899) minus 120573(119872sum119899=1

119901 (119899) minus 1)

minus 120574(119872sum119899=1

2119868119899119872 + 1119901 (119899) minus 2119868119863119905119873119905) (7)




120597119871120597120573 = 119872sum119899=1

119901 (119899) minus 1 = 0 (9)

120597119871120597120574 = 119872sum119899=1

2119868119899119872 + 1119901 (119899) minus 2119868119863119905119873119905 = 0 (10)




2119886 = 119903 (1 minus 119903119872)1 minus 119903 (12)

where

119886 = 1ln 2 + 120573

119903 = 2minus1205742119868(119872+1)(13)


119863119905= 1198731199052minus119886(119872 + 1) (





0

01

02

03

04

05

06

07

08

09

1

Sym

bol p

roba

bilit

y

03 04 05 06 07 0802Dimming ratio

p(1)

p(2)

p(3)

p(4)




119862119889 = max119901(119899)

[119868 (XY1) minus 119868 (XY2)] (15)







(16)


119875119883 (X) = 119872sum119899=1

119901 (119899) 120575 (X minus b119899)

119875119884 (Y1) = 119872sum119899=1

119901 (119899) 119875119885 (Y1 minus b119899)

119875119884 (Y2) = 119872sum119899=1



119879Cminus1119885 Z2)

(17)







0

01

02

03

04

05

06

07Se

crec

y di

mm

ing

capa

city

(bits

sym

bol)

Nt = 1 ( = 04)Nt = 2 ( = 04)Nt = 3 ( = 04)

Nt = 1 ( = 06)Nt = 2 ( = 06)Nt = 3 ( = 06)



0

02

04

06

08

1

12

14

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)


Nt = 1 ( = 04)Nt = 2 ( = 04)Nt = 3 ( = 04)

Nt = 1 ( = 06)Nt = 2 ( = 06)Nt = 3 ( = 06)




0

005

01

015

02

025

03

035

04

045

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)

045 05 055 06 065 07 075 0804Dimming ratio

Nt = 1

Nt = 2

Nt = 3

(a) 120572 = 04


Nt = 1

Nt = 2

Nt = 3

0

005

01

015

02

025

03

035

04

045

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)

045 05 055 06 065 07 075 0804Dimming ratio

(b) 120572 = 06





5 Conclusions





Acknowledgments


References
















RoboticsJournal of





Journal of



RotatingMachinery



Journal of

Volume 201


VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










0

01

02

03

04

05

06

07

08

09

1

Sym

bol p

roba

bilit

y

03 04 05 06 07 0802Dimming ratio

p(1)

p(2)

p(3)

p(4)




119862119889 = max119901(119899)

[119868 (XY1) minus 119868 (XY2)] (15)







(16)


119875119883 (X) = 119872sum119899=1

119901 (119899) 120575 (X minus b119899)

119875119884 (Y1) = 119872sum119899=1

119901 (119899) 119875119885 (Y1 minus b119899)

119875119884 (Y2) = 119872sum119899=1



119879Cminus1119885 Z2)

(17)







0

01

02

03

04

05

06

07Se

crec

y di

mm

ing

capa

city

(bits

sym

bol)

Nt = 1 ( = 04)Nt = 2 ( = 04)Nt = 3 ( = 04)

Nt = 1 ( = 06)Nt = 2 ( = 06)Nt = 3 ( = 06)



0

02

04

06

08

1

12

14

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)


Nt = 1 ( = 04)Nt = 2 ( = 04)Nt = 3 ( = 04)

Nt = 1 ( = 06)Nt = 2 ( = 06)Nt = 3 ( = 06)




0

005

01

015

02

025

03

035

04

045

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)

045 05 055 06 065 07 075 0804Dimming ratio

Nt = 1

Nt = 2

Nt = 3

(a) 120572 = 04


Nt = 1

Nt = 2

Nt = 3

0

005

01

015

02

025

03

035

04

045

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)

045 05 055 06 065 07 075 0804Dimming ratio

(b) 120572 = 06





5 Conclusions





Acknowledgments


References
















RoboticsJournal of





Journal of



RotatingMachinery



Journal of

Volume 201


VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation












0

01

02

03

04

05

06

07Se

crec

y di

mm

ing

capa

city

(bits

sym

bol)

Nt = 1 ( = 04)Nt = 2 ( = 04)Nt = 3 ( = 04)

Nt = 1 ( = 06)Nt = 2 ( = 06)Nt = 3 ( = 06)



0

02

04

06

08

1

12

14

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)


Nt = 1 ( = 04)Nt = 2 ( = 04)Nt = 3 ( = 04)

Nt = 1 ( = 06)Nt = 2 ( = 06)Nt = 3 ( = 06)




0

005

01

015

02

025

03

035

04

045

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)

045 05 055 06 065 07 075 0804Dimming ratio

Nt = 1

Nt = 2

Nt = 3

(a) 120572 = 04


Nt = 1

Nt = 2

Nt = 3

0

005

01

015

02

025

03

035

04

045

Secr

ecy

dim

min

g ca

paci

ty (b

itss

ymbo

l)

045 05 055 06 065 07 075 0804Dimming ratio

(b) 120572 = 06





5 Conclusions





Acknowledgments


References
















RoboticsJournal of





Journal of



RotatingMachinery



Journal of

Volume 201


VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










5 Conclusions





Acknowledgments


References
















RoboticsJournal of





Journal of



RotatingMachinery



Journal of

Volume 201


VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









RoboticsJournal of





Journal of



RotatingMachinery



Journal of

Volume 201


VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









Full-Text PDF - downloads.hindawi.comdownloads.hindawi.com/journals/wcmc/2017/4094096.pdf ·...

Documents

Transcript of Full-Text PDF - downloads.hindawi.comdownloads.hindawi.com/journals/wcmc/2017/4094096.pdf ·...