research article - hindawi publishing...

Research ArticleAn Efficient Contention-Window Based Reporting for Internet ofThings Features in Cognitive Radio Networks

Muhammad Sajjad Khan 12 Junsu Kim 2 Eung Hyuk Lee2 and Su Min Kim 2

1Department of Electrical Engineering International Islamic University Islamabad Islamabad Pakistan2Department of Electronics Engineering Korea Polytechnic University 237 Sangidaehak-ro Siheung-siGyeonggi-do 15073 Republic of Korea

Correspondence should be addressed to Su Min Kim suminkimkpuackr

Received 24 May 2019 Revised 19 July 2019 Accepted 25 July 2019 Published 18 August 2019

Guest Editor Fadi Al-Turjman

Copyright copy 2019 Muhammad Sajjad Khan et al is is an open access article distributed under the Creative CommonsAttribution License which permits unrestricted use distribution and reproduction in anymedium provided the original work isproperly cited

Internet of things (IoT) is a new challenging paradigm for connecting heterogeneous networks However an explosive increase inthe number of IoT cognitive users requires a mass of sensing reporting thus it increases complexity of the system Moreoverbandwidth utilization reporting time and communication overhead arise To realize spectrum sensing how to collect sensingresults by reducing the communication overhead and the reporting time is a problem of major concern in future wirelessnetworks On the other hand cognitive radio is a promising technology to access the spectrum opportunistically In this paper wepropose a contention-window based reporting approach with a sequential fusion mechanism e proposed reporting schemereduces the reporting time and the communication overhead by collecting sensing results from the secondary users with thehighest reliability at a fusion center by utilizing Dempster-Shafer evidence theorye fusion center broadcasts the sensing resultsonce a global decision requirement is satised rough simulations we evaluate the proposed scheme in terms of percentage ofthe number of reporting secondary users error probability percentage of reporting and spectral eciency As a result it is shownthat the proposed scheme is more eective than a conventional order-less sequential reporting scheme

1 Introduction

Wireless communication networks have tremendous prog-ress in the last 30 years to support the growth of the appli-cation devices from 1G to 4G LTE-Advanced wirelessnetwork [1] Each generation has played its role to enhancedata rate reliability latency etc During the past yearsconnecting each device with another device at anytime andanywhere is a big challenge in wireless communicationnetworks In a line of evolution 5G will provide an un-expected contribution and a big step forward toward thespectrum management public safety energy eciency highdata rate low latency and so on [2ndash5]

On the other hands unmanned aerial vehicle (UAV) isexpected to be an important component of the upcomingwireless network ie 5G because of its countless applica-tions such as public safety health management and re-motely services [6 7] In [8] the authors deployed a UAV-

based cognitive system to maximize energy eciency byoptimizing the transmit power Similarly in [9] the authorsstudied resource allocation and trajectory design for anenergy-ecient secure UAV communication system wherea UAV base station serves multiple secondary users in thepresence of the potential eavesdroppers One of the potentialUAV applications is to remotely deploy and monitor sensordevices for future Internet of things (IoT) networks

IoT was rst mentioned by Ashton who introduces atechnological revolution to bring heterogeneous networksunder a single umbrella of the IoT [10] IoT is a promisingsubject of technical social and economic implications it canbe presumed that IoT has a strong andmeaningful impact ondaily life in the near future such as automation improvisedlearning logistic intelligent transportation e-health careand so on [11 12] Technically the most focused area ofparadigm is computing communication and connectivityin IoT Among them the connectivity and management

HindawiWireless Communications and Mobile ComputingVolume 2019 Article ID 8475020 9 pageshttpsdoiorg10115520198475020

spectrum are more challenging and of great concern Ad-ditionally it is noteworthy to mention that with the rapidincrease in connecting devices a lot of spectrum are requiredfor coverage and capacity of these connecting devices in IoTFemtocell can be a promising candidate to meet the demandfor capacity and coverage of the growing IoTdevices [13 14]As over 50 billionwireless devices will be connected by 2020 allof which will demand a lot of spectrum resources [15] theauthors in [16] argued the importance of the cognitive capa-bility that is without comprehensive cognitive capability IoTis just like an awkward stegosaurus all brawn and no brains-e static management and allocation of spectrum resourcesare not efficient to meet requirements of wireless devices andapplications With static allocation of spectrum resourcessome of them are heavily overloaded whereas another part ofthe spectrum is rarely used According to the report revealed byFederal Communication Commission (FCC) the spectrumusage varies from 15 to 85 in some cases [17]

One of tempting solutions against the spectrum shortageis cognitive radio technology (CRT) which has yield ex-tensive studies on spectrum allocation and management fordecades [18] CRT can be integrated with IoT to provide amore intelligent and efficient networking and resourceutilization [19] CRT allows wireless devices to sense thespectrum bands search for suitable frequency channels andreconfigure their parameters to meet channel requirementswhile minimizing energy consumption [20] In CRTspectrum sensing plays a vital role to find a spectrum holeand efficiently utilize the spectrum while avoiding in-terference to primary users (PUs) [21] In [22 23] theauthors have presented a detailed survey for the key enablingtechniques such as detection localization tracking andcontrolling A number of various detection techniques such asenergy detection matched filtering and cyclostationary havebeen utilized to detect the existence of the PU in the network[24 25] Among these techniques the energy detection is oneof the most engaging techniques thanks to its ease ofimplementation and no requirement on prior information ofthe PU However a major drawback of the energy detection isweakness of the received signal strength induced by fadingshadowing and hidden terminal problem

-e fading and hidden terminal problems can beovercome by cooperative spectrum sensing (CSS) [26] CSSmainly consists of three steps sensing reporting and globaldecision In [27] the authors considered a centralized CSSmechanism to make a firm decision for the status of thechannel and broadcast it to the others in the network theauthors also optimized the number of secondary users (SUs)and threshold to get the best result with minimum resourcesIn [28] the authors proposed a novel CoMAC-based CSSscheme that allow cooperative SUs to encode their localstatistics in transmit power and to transmit sequence in-formation of the modulated symbols to a fusion center (FC)for making a final decision on the existence of the PU -eauthors in [29] have improved the sensing performance byadjusting parameters such as decision threshold sensingfrequency and the number of sensing operations

In this paper we propose an ordered-sequential reportingmechanism based on contention window and D-S evidence

theory IoTs in order to reduce the reporting time andcommunication overheads which ultimately reduce costssuch as control channel bandwidth and energy consumptionOnce sensing is performed by utilizing energy detectiontechniques SUs determine its basic probability assignmentand reliability -en SUs wait for listening to the medium(control channel) and contents for the channel access-e SUwith the highest reliability wins the contention and reservetime slots for transmitting sensing reports to the FC-en theFC broadcasts a burst of report messages in the whole me-dium (control channels) if the global requirement is satisfied-rough simulations we demonstrate that the proposedscheme achieves better performance than a conventionalorder-less sequential reporting scheme -e main contribu-tions of this paper are summarized as follows

(i) We propose a contention-window based mecha-nism in which SUs content for the channel access-e SU with the highest reliability wins and accessthe channel for reporting sensing information to theFC

(ii) We propose an ordered-sequential reportingscheme instead of conventional order-less sequen-tial reporting scheme which ultimately enhancesthe performance of the system including reducedreporting time

(iii) To this end we utilize Dempster-Shafer evidencetheory in combining reports at the FC to decide theexistence of PU in the network

(iv) We evaluate the proposed scheme in terms ofpercentage of number of sensing reports errorprobability percentage of reporting and spectralefficiency -rough simulations we demonstratethat the proposed scheme outperforms the order-less sequential reporting scheme

-e remainder of this paper is organized as follows InSection 2 we present related work In Section 3 we discussthe cooperative spectrum sensing and sequential fusion InSection 4 we provide a detailed description of the proposedcontention-window based reporting scheme In Section 5the numerical results are shown Finally the paper is con-cluded in Section 6

2 Related Work

In recent years reporting in CSS has drawn much moreattention In [30] the author proposed a random-accessmechanism in which the author discussed how to collect localsensing reports in CSS A backward induction approach isapplied to decide the optimal stopping time of the collectionperiod Similarly in [31] the authors designed a reportingchannel scheme based on random-access protocols includingslotted ALOHA and reservation ALOHA to measure theperformance of the probability of detection and probability offalse alarm In [32] the author proposed a CSS scheme forcognitive radio networks (CRN) with limited reporting Twokinds of CSS approach with limited reporting in a centralizedCRN have been proposed a soft combination approach with

2 Wireless Communications and Mobile Computing

threshold-based reporting and a soft combination approachwith contention-based reporting In [33] the authorsdesigned a reporting channel structure based on a random-access protocol which is introduced for SUs and FC inaddition k-out-of-N rules are implemented at FC to de-termine the global detection

A data fusion scheme for CSS based on Dempster-Shafer (D-S) theory was first proposed in [34]-is schemehas significantly improved the probability of detection andthe probability of false alarm -e performance of D-Sevidence theory can be enhanced to obtain a larger gain ofcombination by utilizing the signal-to-noise-ratio (SNR)of the PU [35] However the advantages of performanceenhancement cost overhead in traffic control signalingwhich results in consuming more communication re-sources such as reporting time delay control channelbandwidth and transmission energy -e resource de-mands drastically increase with the increasing of the numberof SUs However only few researchers have addressed theseproblems In [36] the authors proposed a sequential test tocontrol the number of reporting bits and average detectiontime Similarly in [37] the authors proposed a cooperativesequential detection scheme to reduce the sensing timeHowever these schemes do not utilize an ordered-sequentialapproach for fast detection with a limited number of reportsOur proposed approach is ordered-sequential contention-window based reporting by the SU with the highest reliabilitywhich significantly reduces reporting time and the errorprobability and after all it is more efficient than a conven-tional order-less sequential reporting scheme

3 Cooperative Sensing and Sequential Fusion

We consider a cooperative sensing scenario in a CRN in whicheach SU conducts local sensing by utilizing energy detectiontechnique and collects basic information about the PU status inthe network After local sensing each SU measures its basicprobability assignment (BPA) and reliability and reports thisinformation to FC Figure 1 shows the cooperative sensingscenario and a frame structure consisting of sensing periodreporting period and transmission

At the beginning each SU performs local spectrumsensing in a distributed manner -e local spectrum sensingcan be represented as a binary hypothesis testing of thepresence or absence of PU in the network and is measured as

yi(n) z(n) H0

hi(n)s(n) + z(n) H11113896 (1)

where H0 is the absence H1 is the presence of the PU hi(n)

is the fading coefficient z(n) is the additive white Gaussiannoise (AWGN) s(n) is the PU signal and yi(n) is the signalreceived at the i-th SU respectively

Each SU measures test statistics of the received signal byutilizing an energy detection technique given as

XEi(n) 1113944

NT

k1yk(n)

111386811138681113868111386811138681113868111386811138682 (2)

where NT 2TW in which T is the sensing time and W isthe bandwidth and yk is the j-th sample of the receivedsignal When NT is large enough by central limit theorem(CLT) XEi can be well approximated as a Gaussian distri-bution [26]

After performing local sensing each SU measures itsself-assessed credibility which is equivalent to BPA for H0and H1 hypotheses BPA is defined as a form of cumulativedistribution function given as [35]

H0 mi H0( 1113857 1113946infin

XEi

12πσ0i

eminus XEi minus μ0i( )

2σ20i dx

H1 mi H1( 1113857 1113946infin

XEi

12πσ0i


2σ21i dx

⎧⎪⎪⎪⎪⎪⎪⎨

⎪⎪⎪⎪⎪⎪⎩

(3)

where mi(H0) and mi(H1) are the BPAs of hypothesis H0the absence of PU and hypothesis H1 the presence of PU

According to D-S evidence theory the BPAs of each SUat the FC are combined as

m H0( 1113857 1113936A1capA1capANH0

1113937Ni1mHi Ai( 1113857

1 minus K

m H1( 1113857 1113936A1capA1capANH1

1113937Ni1mHi Ai( 1113857

1 minus K

whereK 1113944A1capA1capANempty

1113937N

i1mHi Ai( 1113857

⎧⎪⎪⎪⎪⎪⎪⎪⎪⎪⎨

⎪⎪⎪⎪⎪⎪⎪⎪⎪⎩

(4)

Sensing

Sensing period

Time slot (T)

Transmission

TxTs

Reporting period

SU detection SU reporting

PU

Sensing stage

SU1

SU2

SU3SUM

FCReporting stage

Figure 1 System model

Wireless Communications and Mobile Computing 3

where (Ai) is one of the elements of a set H0 H1Ω1113864 1113865 inwhich Ω is the ignorance hypothesis that either hypothesisH0 H11113864 1113865 can be true

-e main problem of D-S evidence theory as well asother schemes is that it requires a large amount of resourcesfor reporting sensing results

In order to reduce the overhead processing time andreporting time and energy consumption we consider anordered-sequential fusion CSS After local sensing each SUmeasures its BPA At FC the BPAs are combined sequen-tially according to their reliability

It is noteworthy that the combined results of the pro-posed ordered-sequential fusion are equal to nonsequentialone when all SUs send reports to FC -erefore instead ofkeeping threshold 0 we adopt a threshold (η) -e thresholdis set to a large enough value so that the cooperation gain isequivalently maintained even if the number of combinedsensing results is lower

-e final sequential decision is based on the followingstrategy

-e following two strategies are applied to FC to make aglobal decision When the number of reports k at the FC isless than the total number of SUs the global decision can bedetermined as

Fd

H0 rkglobal lt minus η

H1 rkglobal gt η

no decision minus ηlt rkglobal lt η

⎧⎪⎪⎪⎨

⎪⎪⎪⎩

(5)

where the condition minus ηlt rkglobal lt η denotes that the reportsat the FC are not enough to declare a global decision andwait for more sensing reports rkglobal represents the globaldecision reliability at the k minus th report expressed as

rkglobal mkglobal H1( 1113857

mkglobal H0( 11138571113888 1113889 (6)

where mkglobal(Hj) mkminus 1global(Hj)oplusmkglobal(H j) is theorder of sequential combination of the BPAs j [0 1] andmkglobal(Hj) and mkminus 1global(Hj) are the k-th and (k minus 1)-thglobal BPA hypotheses Hj respectively

When the number of reports k is equal to the totalnumber of SU at the FC the global decision can be de-termined as

Fd H0 rkglobal lt 0

H1 rkglobal gt 0

⎧⎨

⎩ (7)

4 Proposed Contention-Window BasedReporting Scheme

In this section we first describe the conventional order-lesssequential fusion reporting scheme and then discuss theproposed contention-window based reporting scheme

At the beginning each SU performs spectrum sensingmeasures its BPA information and waits for the FC requestto send their information to the FC

In the conventional order-less sequential fusion reportingscheme the FC sends a request to SUs in a predefined order

-en the requested SU responds its own sensing report to theFC-e FC accumulates the sensing reports of the current SUwith the previous SU and verifies whether the decision re-quirement is satisfied or not If the required decision issatisfied the FC sends the stop reporting message to all SUsand further reporting is stopped If the decision is not sat-isfied FC sends a request to next SU in the predefined orderand the process is repeated until the decision requirement issatisfied -e overall process for the conventional order-lesssequential reporting scheme is shown in Figure 2

In the proposed reporting scheme instead of reportingin a predefined manner we consider a contention-windowbased reporting scheme in which the SUs report to the FCaccording to their sensing data reliability given in (6) -esensing data reliability of each SU depends on the offset timefor accessing the medium (control channel) for reporting tothe FC for its turn As in this mechanism SUs content toaccess medium (control channel) for a specific time to reportits sensing information to the FC thus it is named as thecontention-window based reporting scheme

In the reporting period from the beginning of the con-tention slot each SU listens to themedium (control channel) Ifthere is no signal till toffset for SUs then it is assumed that SUwins the contention slot tcon -e winner SU generates a burstsignal to the medium (control channel) for reservation of thechannel and reports in the next reporting slot It is worthnoting that the offset is in the range of tslot lt toffset lt ttcon

minus tslot-e first tslot is reserved for the SU with higher priority for thetransmission to the FC Whenever the global decision re-quirement is satisfied the FC sends a burst message signal inthe whole medium (control channel) -e overall proposedcontention-based reporting scheme is shown in Figure 3

In the k-th contention slot the value of the offset timetoffset is calculated as follows

toffset tcon minus 2tslot

rkmax minus rkmin1113872 1113873rkmax minus ri1113872 1113873 + tslot (8)

where tslot is the slot time corresponding to the time requiredby the radio layer carrier sensing to function and tcon is thelength of the contention slot rkmax and rkmin are themaximum and minimum values of the data sensing re-liability at the k-th contention slot respectively -e valuesof rkmax and rkmin are first time defined by the FC and thenupdated automatically after every reporting slot

-e average reporting time Tavg_rep for the order-lesssequential reporting scheme can be determined as

Tavg_rep M middot Ppoll middot treport + tpoll1113872 1113873 (9)

where M is the number of SUs and Ppoll is the percentage ofthe number of reporting SUs

-e average reporting time for the proposed contention-window based reporting scheme Tavg_rep_prop is determined as

Tavg_rep_prop M middot Pcon middot treport + tcon1113872 1113873 (10)

where Pcon is the percentage of the number of reporting SUsfor the proposed contention-window based reporting scheme

-e overall flowchart of the proposed contention-win-dow based reporting scheme is shown in Figure 4


5 Numerical Evaluation

In this section we present simulation results for the pro-posed contention-window based reporting scheme in

consideration with the ordered-sequential fusion andcompare its performance with the conventional order-lesssequential reporting scheme We consider IEEE 80222standards it is assumed that the existence of the PU is 05

Sensing period

Reporting period

treport

Transmission period

FC definesreliability range

at begining

FC burst

Global decisionbroadcast if the finaldecision is satisfied

Contention slot (tcon)

tslottslotSU burst

toffset

Figure 3 Proposed contention-window based reporting

FC sends a request toSUs in a predefined

order

The requested SU sendssensing report to FC

FC collects reports fromSUs

FCaccumulates and verifies

decision criteriasatisfied

No

Yes

All SUs stop furtherreporting to FC and broadcast

the global decision

Move to the next SU forcollecting report

Figure 2 Conventional order-less sequential fusion reporting scheme


and the used bandwidth is 6MHz -e simulation envi-ronment is developed by utilizing MATLAB as an imple-mentation tool -e parameters for the numerical evaluationare summarized in Table 1

Figure 5 shows the percentage of the number of reportingSUs against the different value of SNRs (minus 20 to minus 10dB) It canbe observed from Figure 5 that as the number of SUs in thenetwork increases the percentage of the number of reportingSUs decreases for the proposed contention-based ordered-

sequential fusion for a specific threshold value (thresholdη 15) Specifically it is shown in Figure 5 that with the order-less sequential reporting scheme approximately 37 of thenumber of reporting SUs is required to satisfy the global de-cision requirement whereas the proposed scheme requiresapproximately 14 of the number of reporting SUs when 100SUs are considered in the network -e reason is obvious thatinstead of the SU reporting in an order-less manner only SUwith the highest reliability reports to FC to satisfy the globaldecision requirement which ultimately reduces the number ofreports -e proposed contention-window based reportingscheme requires less percentage of the number of reportscompared with the order-less sequential reporting scheme It isnoted that as the SNR value increases the reports for theproposed scheme decrease Also it is worth noting that as thevalue of the threshold increases the percentage of the number ofreporting SUs increases

Figure 6 shows the performance comparison of the pro-posed scheme in terms of error probability for varying averagenumber of reporting SUs It can be observed from Figure 6 thaterror probability of the proposed contention-window basedreporting scheme is smaller than the order-less sequentialreporting scheme and it becomes identical to that of the order-less scheme as the average number of reporting SUs increases Itis obvious that in the beginning of the proposed contention-window based reporting the highly reliable SUs report to FCthus its global requirements converge and the error probability

Radio environment

BPAmass valuemeasurement by each

SU

SU measures itsreliability values

FC defines reliabilityranges at the begining

SU contents to accessthe control channel

Winner SU bursts inchannel for reporting in

the next slot


FC global decisionsatisfied

SU updates reliabilityrange

FC bursts global decisionto all SUs

Yes

No

Figure 4 Flow chart for the proposed contention-window basedreporting scheme

Table 1 Simulation parameters

Parameter ValueNumber of SUs 100 150 200Probability of PU appearance 05Bandwidth 6MHzSlot time tslot 20 μsecPoll time tpoll 200 μsecContention slot tcon 200 μsecReporting slot interval treport 1msec

ndash20 ndash19 ndash18 ndash17 ndash16 ndash15 ndash14 ndash13 ndash12 ndash11 ndash10

40

35

30

25

20

15

10

5

0Pe

rcen

tage

of t

he n

umbe

r of

repo

rtin

g SU

s

Proposed reportingscheme SU = 100Proposed reportingscheme SU = 150Proposed reportingscheme SU = 200

Orderless reportingscheme SU = 100Orderless reportingscheme SU = 150Orderless reportingscheme SU = 200

SNR

Figure 5 Percentage of the number of reporting SUs vs SNR


is smaller than the order-less sequential fusion scheme How-ever as the average number of reporting SUs increases moreSUs will report to FC to meet the required global decision andconverges and thus the error probability decreases As a resultat a large average number of reporting SUs the error probabilityis low and almost the identical for both schemes

Figure 7 shows the percentage of reporting for varyingthreshold value It is worth noting that the proposed con-tention-window based reporting scheme requires a smallernumber of SUs for reporting than the conventional order-less reporting scheme Also it is important to mention thatas the number of threshold value increases the percentage ofreporting increases It is well justified because more SUsreport to FC in order to satisfy the global decision re-quirement After all it is shown that the proposed con-tention-window based reporting scheme is more effectivethan the conventional order-less sequential fusion reportingscheme

Figure 8 shows the spectral efficiency for varying de-tection probability As the detection probability increasesspectrum availability for the SUs decreases thus the spectral

efficiency of the system decreases However the proposedscheme achieves better performance than the conventionalorder-less scheme As a result the proposed scheme is alsomore effective in the perspective of spectral efficiency

6 Conclusion

-e integration of cognitive radio technology and the In-ternet of things seems to shift future wireless networks (5G)Cognitive radio technology has the potential to efficientlyutilize the spectrum via cooperative sensing However therise in cooperative sensing users increases the averagereporting time bandwidth utilization and communicationoverhead In this paper we proposed an ordered-sequentialfusion scheme with contention-window based reporting forInternet of things to reduce the reporting time and thecommunication overhead which ultimately reduces costand bandwidth utilization Secondary users content to accessthe shared control channel and report their information toFC based on reliability to reduce the reporting time durationby satisfying the global decision requirement -e effec-tiveness of the proposed contention-window based report-ing scheme is shown through simulations by consideringpercentage of the number of reporting SUs error proba-bility percentage of reporting and spectral efficiency -eresults showed that the proposed ordered-sequential con-tention-window based reporting scheme outperforms theconventional order-less sequential reporting scheme invarious aspects

Data Availability

-e data used to support the findings of this study are in-cluded within the article

Conflicts of Interest

-e authors declare no conflicts of interest

Erro

r pro

babi

lity

05

045

04

035

03

025

02

015

Averge number of reporting SUs0 10 20 30 40 50 60 70 80 90 100

SNR = ndash20

Proposed ordered schemeOrderless scheme

Figure 6 Error probability vs average number of reporting SUs

100908070605040302010

00 10 20 30 40 50 60 70

reshold value

Perc

enta

ge o

f rep

ortin

g

SNR = ndash20


Figure 7 Percentage of reporting vs threshold value

14

12

1

08

06

04

02

007 075 08 085 09

Detection probability

Spec

tral

effic

ienc

y


Figure 8 Spectral efficiency vs detection probability


Acknowledgments

-is work was supported in part by the MSIT (Ministry ofScience and ICT) Korea under the ITRC (InformationTechnology Research Center) support program (IITP-2019-2018-0-01426) supervised by the IITP (Institute for In-formation and Communication Technology Planning ampEvaluation) and in part by the National Research Founda-tion (NRF) funded by the Korean government (MSIT) (No2019R1F1A1059125)

References

[1] A Agarwal G Misra and K Agarwal ldquo-e 5th generationmobile networks-key concepts network architecture andchallengesrdquo American Journal of Electrical and ElectronicEngineering vol 3 no 2 pp 22ndash28 2015

[2] H Shakhatreh A Sawalmeh A Al-Fuqaha et al UnmannedAerial Vehicles A Survey on Civil Applications and Key Re-search Challenges IEEE Access Piscataway NJ USA 2019

[3] Z Kaleem and M H Rehmani ldquoAmateur drone monitoringstate-of-the-art architectures key enabling technologies andfuture Research directionsrdquo IEEE Wireless Communicationsvol 25 no 2 pp 150ndash159 2018

[4] M A Khan I M Qureshi and F Khanzada ldquoA hybridcommunication scheme for efficient and low-cost deploymentof future flying ad-hoc network (FANET)rdquo Drones vol 3no 1 p 16 2019

[5] Z Kaleem Y Li and K Chang ldquoPublic safety usersrsquo priority-based energy and time-efficient device discovery scheme withcontention resolution for ProSe in third generation part-nership project long-term evolution-advanced systemsrdquo IETCommunications vol 10 no 15 pp 1873ndash1883 2016

[6] Z Kaleem M H Rehmani E Ahmed et al ldquoAmateur dronesurveillance applications architectures enabling technolo-gies and public safety issues part 1rdquo IEEE CommunicationsMagazine vol 56 no 1 pp 14-15 2018

[7] Z Kaleem M Yousaf A Qamar et al ldquoUAV-empowereddisaster-resilient edge architecture for delay-sensitive com-municationrdquo 2019 httpsarxivorgabs180909617v2

[8] L Sboui H Ghazzai Z Rezki and M Alouini ldquoEnergy-Efficient power allocation for UAV cognitive radio systemsrdquoin Proceedings of the IEEE 86th Vehicular Technology Con-ference (VTC-Fall) IEEE Toronto ON Canada September2017

[9] Y Cai Z Wei R Li D W K Ng and J Yuan ldquoEnergy-efficient resource allocation for secure UAV communicationsystemsrdquo 2019 httparxivorgabs190109308

[10] K Ashton ldquo-at ldquointernet of thingsrdquo in the real world thingsmatter more than ideasrdquo RFID Journal vol 22 2009

[11] S Chatterjee R Mukherjee S Ghosh D Gosh S Gosh andA Mukherjee ldquoInternet of things and cognitive radio-issuesand challengesrdquo in Proceedings of the IEEE InternationalConference on Opto-Electronics and Applied Optics (Optronix)IEEE Kolkata India November 2017

[12] A A Khan M H Rehmani and A Rachedi ldquoWhen cognitiveradio meets the internet of thingsrdquo in Proceedings of the IEEEInternational Wireless Communications and ComputingConference (IWCMC) IEEE Paphos Cyprus September2016

[13] F Al-Turjman E Ever and H Zahmatkesh ldquoGreen femto-cells in the IoT era traffic modeling and challengesmdashanoverviewrdquo IEEE Network vol 31 no 6 pp 48ndash55 2017

[14] F Al-Turjman E Ever and H Zahmatkesh ldquoSmall cells in theforthcoming 5GIoT traffic modeling and deploymentoverviewrdquo IEEE Communications Surveys and Tutorialsvol 21 no 1 pp 28ndash65 2018

[15] A Al-Fuqaha M Guizani M Mohammadi M Aledhari andM Ayyash ldquoInternet of things a survey on enabling tech-nologies protocols and applicationsrdquo IEEE CommunicationsSurveys and Tutorials vol 17 no 4 pp 2347ndash2376 2015

[16] Q Wu G Ding Y Xu et al ldquoCognitive internet of things anew paradigm beyond connectionrdquo IEEE Internet of ingsJournal vol 1 no 2 pp 129ndash143 2014

[17] FCC ldquoNotice of proposed rule-making and orderrdquo ET DocketNo 03-222 FCC Washington DC USA 2003

[18] P Rawat K D Singh and J M Bonnin ldquoCognitive radio forM2M and internet of things a surveyrdquo Computer Commu-nications vol 94 pp 1ndash29 2016

[19] S Aslam W Ejaz and M Ibnkahla ldquoEnergy and spectralefficient cognitive radio sensor networks for internet ofthingsrdquo IEEE Internet of ings Journal vol 5 no 4pp 3220ndash3233 2018

[20] J Mitola and G Q Maguire ldquoCognitive radio makingsoftware radio more personalrdquo IEEE Personal Communica-tion vol 6 no 4 pp 13ndash18 1999

[21] Y Arjoune and N Kaabouch ldquoA comprehensive survey onspectrum sensing in cognitive radio netowks recent advancesnew challenges and future research directionrdquo Sensorsvol 19 no 1 p 126 2019

[22] G Ding Q Wu L Zhang Y Lin T A Tsiftsis and Y YaoldquoAn amateur drone surveillance system based on cognitiveinternet of thingsrdquo IEEE Communications Magazine vol 56no 1 pp 29ndash35 2018

[23] Y Saleem M H Rehmani and S Zeadally ldquoIntegration ofcognitive radio technology with unmanned aerial vehiclesissues opportunities and future research challengesrdquo Journalof Network and Computer Applications vol 50 pp 15ndash312015

[24] A Ranjan Anurag and B Singh ldquoDesign and analysis ofspectrum sensing in cognitive radio based on energy de-tectionrdquo in Proceedings of the IEEE International Conferenceon Signal and Information Processing (IConSIP) IEEENanded India October 2016

[25] I Ilyas S Paul A Rahman and R K Kundu ldquoComparativeevaluation of cyclo-stationary detection based cognitivespectrum sensingrdquo in Proceedings of the IEEE 7th AnnualUbiquitous Computing Electronics and Mobile Communica-tion Conference (UEMCON) IEEE New York NY USAOctober 2016

[26] I F Akilidz B F Lo and R Balakrishnan ldquoCooperativespectrum sensing in cognitive radio networks a surveyrdquoPhysical Communication vol 4 no 1 pp 40ndash62 2011

[27] S C Shinde and A N Jadhav ldquoCentralized cooperativespectrum sensing with energy detection in cognitive radio andoptimizationrdquo in Proceedings of the IEEE InternationalConference on Recent Trends in Electronics Information andCommunication Technology (RTEICT) IEEE Bangalore In-dia May 2016

[28] M Zheng C Xu W Liang H Yu and L Chin ldquoA novelCoMAC based cooperative spectrum sensing scheme incognitive radio networksrdquo in Proceedings of the IEEE In-ternational Conference on Communication Workshop(ICCW) IEEE London UK June 2015

[29] H Hu Y Xu Z Liu N Li and H Zhang ldquoOptimal strategiesfor cooperative spectrum sensing in multiple cross over


cognitive radio networksrdquo KSII Transaction on Internet andInformation System vol 6 no 12 pp 3061ndash3080 2012

[30] D Lee ldquoAdaptive random access for cooperative spectrumsensing in cognitive radio networkrdquo IEEE Transaction onWireless Communications vol 14 no 2 pp 831ndash840 2015

[31] R Alhamad H Wang and Y Yao ldquoCooperative spectrumsensing with random access reporting channels in cognitiveradio networksrdquo IEEE Transaction on Vehicular Technologyvol 66 no 8 pp 7249ndash7261 2017

[32] J So ldquoCooperative spectrum sensing for cognitive radionetworks with limited reportingrdquo KSII Transaction on In-ternet and Information Systems vol 9 no 8 pp 2755ndash27732015

[33] R Alhamad H Wang and Y Yao ldquoReporting channel designand analysis in cooperative spectrum sensing for cognitiveradio networksrdquo in Proceedings of the IEEE 82nd VehicularTechnology Conference (VTC-2015-Fall) IEEE Boston MAUSA September 2015

[34] Q Peng K Zeng W Jun and S Li ldquoA distributed spectrumsensing scheme based on credibility and evidence theory incognitive radio networksrdquo in Proceedings of the 17th IEEEInternational Symposium on Personal Indoor and MobileRadio Communication pp 1ndash5 IEEE Helsinki FinlandSeptember 2006

[35] N Nguyen--anh and I Koo ldquoAn enhanced cooperativespectrum sensing scheme based on evidence theory and re-liability source evaluation in cognitive radio contextrdquo IEEECommunications Letters vol 13 no 7 pp 492ndash494 2009

[36] S Yeelin and Y T Su ldquoA sequential test based cooperativespectrum sensing scheme for cognitive radiosrdquo in Proceedingsof the IEEE 19th International Symposium on Personal Indoorand Mobile Radio Communications pp 1ndash5 IEEE CannesFrance September 2008

[37] Z Qiyue Z Songfeng and A H Sayed ldquoCooperative spec-trum sensing via sequential detection for cognitive radionetworksrdquo in Proceedings of the IEEE 10th Workshop onSignal Processing Advances in Wireless Communicationpp 121ndash125 IEEE Perugia Italy June 2009


International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018


Active and Passive Electronic Components

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of



Journal ofEngineeringVolume 2018

SensorsJournal of



RotatingMachinery


Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation


Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

spectrum are more challenging and of great concern Ad-ditionally it is noteworthy to mention that with the rapidincrease in connecting devices a lot of spectrum are requiredfor coverage and capacity of these connecting devices in IoTFemtocell can be a promising candidate to meet the demandfor capacity and coverage of the growing IoTdevices [13 14]As over 50 billionwireless devices will be connected by 2020 allof which will demand a lot of spectrum resources [15] theauthors in [16] argued the importance of the cognitive capa-bility that is without comprehensive cognitive capability IoTis just like an awkward stegosaurus all brawn and no brains-e static management and allocation of spectrum resourcesare not efficient to meet requirements of wireless devices andapplications With static allocation of spectrum resourcessome of them are heavily overloaded whereas another part ofthe spectrum is rarely used According to the report revealed byFederal Communication Commission (FCC) the spectrumusage varies from 15 to 85 in some cases [17]

One of tempting solutions against the spectrum shortageis cognitive radio technology (CRT) which has yield ex-tensive studies on spectrum allocation and management fordecades [18] CRT can be integrated with IoT to provide amore intelligent and efficient networking and resourceutilization [19] CRT allows wireless devices to sense thespectrum bands search for suitable frequency channels andreconfigure their parameters to meet channel requirementswhile minimizing energy consumption [20] In CRTspectrum sensing plays a vital role to find a spectrum holeand efficiently utilize the spectrum while avoiding in-terference to primary users (PUs) [21] In [22 23] theauthors have presented a detailed survey for the key enablingtechniques such as detection localization tracking andcontrolling A number of various detection techniques such asenergy detection matched filtering and cyclostationary havebeen utilized to detect the existence of the PU in the network[24 25] Among these techniques the energy detection is oneof the most engaging techniques thanks to its ease ofimplementation and no requirement on prior information ofthe PU However a major drawback of the energy detection isweakness of the received signal strength induced by fadingshadowing and hidden terminal problem

-e fading and hidden terminal problems can beovercome by cooperative spectrum sensing (CSS) [26] CSSmainly consists of three steps sensing reporting and globaldecision In [27] the authors considered a centralized CSSmechanism to make a firm decision for the status of thechannel and broadcast it to the others in the network theauthors also optimized the number of secondary users (SUs)and threshold to get the best result with minimum resourcesIn [28] the authors proposed a novel CoMAC-based CSSscheme that allow cooperative SUs to encode their localstatistics in transmit power and to transmit sequence in-formation of the modulated symbols to a fusion center (FC)for making a final decision on the existence of the PU -eauthors in [29] have improved the sensing performance byadjusting parameters such as decision threshold sensingfrequency and the number of sensing operations

In this paper we propose an ordered-sequential reportingmechanism based on contention window and D-S evidence

theory IoTs in order to reduce the reporting time andcommunication overheads which ultimately reduce costssuch as control channel bandwidth and energy consumptionOnce sensing is performed by utilizing energy detectiontechniques SUs determine its basic probability assignmentand reliability -en SUs wait for listening to the medium(control channel) and contents for the channel access-e SUwith the highest reliability wins the contention and reservetime slots for transmitting sensing reports to the FC-en theFC broadcasts a burst of report messages in the whole me-dium (control channels) if the global requirement is satisfied-rough simulations we demonstrate that the proposedscheme achieves better performance than a conventionalorder-less sequential reporting scheme -e main contribu-tions of this paper are summarized as follows

(i) We propose a contention-window based mecha-nism in which SUs content for the channel access-e SU with the highest reliability wins and accessthe channel for reporting sensing information to theFC

(ii) We propose an ordered-sequential reportingscheme instead of conventional order-less sequen-tial reporting scheme which ultimately enhancesthe performance of the system including reducedreporting time

(iii) To this end we utilize Dempster-Shafer evidencetheory in combining reports at the FC to decide theexistence of PU in the network

(iv) We evaluate the proposed scheme in terms ofpercentage of number of sensing reports errorprobability percentage of reporting and spectralefficiency -rough simulations we demonstratethat the proposed scheme outperforms the order-less sequential reporting scheme

-e remainder of this paper is organized as follows InSection 2 we present related work In Section 3 we discussthe cooperative spectrum sensing and sequential fusion InSection 4 we provide a detailed description of the proposedcontention-window based reporting scheme In Section 5the numerical results are shown Finally the paper is con-cluded in Section 6

2 Related Work

In recent years reporting in CSS has drawn much moreattention In [30] the author proposed a random-accessmechanism in which the author discussed how to collect localsensing reports in CSS A backward induction approach isapplied to decide the optimal stopping time of the collectionperiod Similarly in [31] the authors designed a reportingchannel scheme based on random-access protocols includingslotted ALOHA and reservation ALOHA to measure theperformance of the probability of detection and probability offalse alarm In [32] the author proposed a CSS scheme forcognitive radio networks (CRN) with limited reporting Twokinds of CSS approach with limited reporting in a centralizedCRN have been proposed a soft combination approach with







yi(n) z(n) H0

hi(n)s(n) + z(n) H11113896 (1)




XEi(n) 1113944

NT

k1yk(n)

111386811138681113868111386811138681113868111386811138682 (2)



H0 mi H0( 1113857 1113946infin

XEi

12πσ0i


2σ20i dx

H1 mi H1( 1113857 1113946infin

XEi

12πσ0i


2σ21i dx

⎧⎪⎪⎪⎪⎪⎪⎨

⎪⎪⎪⎪⎪⎪⎩

(3)



m H0( 1113857 1113936A1capA1capANH0

1113937Ni1mHi Ai( 1113857

1 minus K

m H1( 1113857 1113936A1capA1capANH1

1113937Ni1mHi Ai( 1113857

1 minus K


1113937N

i1mHi Ai( 1113857

⎧⎪⎪⎪⎪⎪⎪⎪⎪⎪⎨

⎪⎪⎪⎪⎪⎪⎪⎪⎪⎩

(4)

Sensing

Sensing period

Time slot (T)

Transmission

TxTs

Reporting period


PU

Sensing stage

SU1

SU2

SU3SUM

FCReporting stage









Fd


H1 rkglobal gt η


⎧⎪⎪⎪⎨

⎪⎪⎪⎩

(5)



mkglobal H0( 11138571113888 1113889 (6)



Fd H0 rkglobal lt 0

H1 rkglobal gt 0

⎧⎨

⎩ (7)
























Sensing period

Reporting period

treport

Transmission period


at begining

FC burst



tslottslotSU burst

toffset



order





No

Yes


the global decision








Radio environment


SU





the next slot





Yes

No





40

35

30

25

20

15

10

5

0Pe

rcen

tage

of t

he n

umbe

r of

repo

rtin

g SU

s



SNR







6 Conclusion


Data Availability




Erro

r pro

babi

lity

05

045

04

035

03

025

02

015


SNR = ndash20



100908070605040302010

00 10 20 30 40 50 60 70

reshold value

Perc

enta

ge o

f rep

ortin

g

SNR = ndash20



14

12

1

08

06

04

02

007 075 08 085 09


Spec

tral

effic

ienc

y




Acknowledgments


References











































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia







yi(n) z(n) H0

hi(n)s(n) + z(n) H11113896 (1)




XEi(n) 1113944

NT

k1yk(n)

111386811138681113868111386811138681113868111386811138682 (2)



H0 mi H0( 1113857 1113946infin

XEi

12πσ0i


2σ20i dx

H1 mi H1( 1113857 1113946infin

XEi

12πσ0i


2σ21i dx

⎧⎪⎪⎪⎪⎪⎪⎨

⎪⎪⎪⎪⎪⎪⎩

(3)



m H0( 1113857 1113936A1capA1capANH0

1113937Ni1mHi Ai( 1113857

1 minus K

m H1( 1113857 1113936A1capA1capANH1

1113937Ni1mHi Ai( 1113857

1 minus K


1113937N

i1mHi Ai( 1113857

⎧⎪⎪⎪⎪⎪⎪⎪⎪⎪⎨

⎪⎪⎪⎪⎪⎪⎪⎪⎪⎩

(4)

Sensing

Sensing period

Time slot (T)

Transmission

TxTs

Reporting period


PU

Sensing stage

SU1

SU2

SU3SUM

FCReporting stage









Fd


H1 rkglobal gt η


⎧⎪⎪⎪⎨

⎪⎪⎪⎩

(5)



mkglobal H0( 11138571113888 1113889 (6)



Fd H0 rkglobal lt 0

H1 rkglobal gt 0

⎧⎨

⎩ (7)
























Sensing period

Reporting period

treport

Transmission period


at begining

FC burst



tslottslotSU burst

toffset



order





No

Yes


the global decision








Radio environment


SU





the next slot





Yes

No





40

35

30

25

20

15

10

5

0Pe

rcen

tage

of t

he n

umbe

r of

repo

rtin

g SU

s



SNR







6 Conclusion


Data Availability




Erro

r pro

babi

lity

05

045

04

035

03

025

02

015


SNR = ndash20



100908070605040302010

00 10 20 30 40 50 60 70

reshold value

Perc

enta

ge o

f rep

ortin

g

SNR = ndash20



14

12

1

08

06

04

02

007 075 08 085 09


Spec

tral

effic

ienc

y




Acknowledgments


References











































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia








Fd


H1 rkglobal gt η


⎧⎪⎪⎪⎨

⎪⎪⎪⎩

(5)



mkglobal H0( 11138571113888 1113889 (6)



Fd H0 rkglobal lt 0

H1 rkglobal gt 0

⎧⎨

⎩ (7)
























Sensing period

Reporting period

treport

Transmission period


at begining

FC burst



tslottslotSU burst

toffset



order





No

Yes


the global decision








Radio environment


SU





the next slot





Yes

No





40

35

30

25

20

15

10

5

0Pe

rcen

tage

of t

he n

umbe

r of

repo

rtin

g SU

s



SNR







6 Conclusion


Data Availability




Erro

r pro

babi

lity

05

045

04

035

03

025

02

015


SNR = ndash20



100908070605040302010

00 10 20 30 40 50 60 70

reshold value

Perc

enta

ge o

f rep

ortin

g

SNR = ndash20



14

12

1

08

06

04

02

007 075 08 085 09


Spec

tral

effic

ienc

y




Acknowledgments


References











































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia





Sensing period

Reporting period

treport

Transmission period


at begining

FC burst



tslottslotSU burst

toffset



order





No

Yes


the global decision








Radio environment


SU





the next slot





Yes

No





40

35

30

25

20

15

10

5

0Pe

rcen

tage

of t

he n

umbe

r of

repo

rtin

g SU

s



SNR







6 Conclusion


Data Availability




Erro

r pro

babi

lity

05

045

04

035

03

025

02

015


SNR = ndash20



100908070605040302010

00 10 20 30 40 50 60 70

reshold value

Perc

enta

ge o

f rep

ortin

g

SNR = ndash20



14

12

1

08

06

04

02

007 075 08 085 09


Spec

tral

effic

ienc

y




Acknowledgments


References











































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia






Radio environment


SU





the next slot





Yes

No





40

35

30

25

20

15

10

5

0Pe

rcen

tage

of t

he n

umbe

r of

repo

rtin

g SU

s



SNR







6 Conclusion


Data Availability




Erro

r pro

babi

lity

05

045

04

035

03

025

02

015


SNR = ndash20



100908070605040302010

00 10 20 30 40 50 60 70

reshold value

Perc

enta

ge o

f rep

ortin

g

SNR = ndash20



14

12

1

08

06

04

02

007 075 08 085 09


Spec

tral

effic

ienc

y




Acknowledgments


References











































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia






6 Conclusion


Data Availability




Erro

r pro

babi

lity

05

045

04

035

03

025

02

015


SNR = ndash20



100908070605040302010

00 10 20 30 40 50 60 70

reshold value

Perc

enta

ge o

f rep

ortin

g

SNR = ndash20



14

12

1

08

06

04

02

007 075 08 085 09


Spec

tral

effic

ienc

y




Acknowledgments


References











































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia


Acknowledgments


References











































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia














RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia


research article - hindawi publishing...

Documents