B.ParameshwariAssistant ProfessorDepartment ofElectronics andCommunication Engineering,CJITS, Yeshwanthapur,T.
SE-mail:[email protected] paper presents, a spectrum sharing strategy incooperative cognitive radio network (CCRN). A multi-phasecooperation architecture is explained and studied withcooperation partner selection and spectrum sharing amongsecondary users (SUs).The data of primary users (PUs)forwarded to the cooperation partners who are selected fromSUs, and then acquire the spectrum access opportunities fortheir own transmissions as a reward. The partner selection ismodeled as an optimally weighted bipartite matching problemto maximize the total utility where energy efficiency is alsoconsidered just to increase the utility for the PU-SUcooperation pairs.
By the partner SU further improvisation inthe spectrum utilization is done by sharing the acquiredspectrum with the surrounding SUs via cooperative networkcoding. At the end the simulation results provided, whichshows that to the dynamic traffic loads in CCRN, theproposed partner selection and spectrum sharing approachadapts well.KeywordsCCRN, QoS, IUs, MMSE1. INTRODUCTIONThe scarcity of spectral resources has become a severeproblem due to the significant growth in commercial wirelessservices, in recent years, with the emergence of cooperativecommunications in wireless networks 3, a newcommunication paradigm in cooperative cognitive radionetworks is proposed 4–6, termed cooperative cognitiveradio networks (CCRN). The traditional fixed spectrumallocation is proved inefficient, since the frequency band islargely under-utilized 1.
Cognitive Radio (CR) 2 has beenconsidered as a promising technology for improve spectrumutilization by allowing secondary users (SUs) to accessspectrum holes unoccupied by primary users (Pus).The rapid growth in wireless communications has contributeda huge demand on the deployment of new wireless services inboth the licensed and unlicensed frequency spectrum.However, recent Studies show the fixed spectrum assignmentpolicy enforced today results in poor spectrum utilization. Toaddress this problem, cognitive radio 1,2 has emerged as apromising technology to enable the access of the intermittentperiods of unoccupied frequency bands, known as white spaceor spectrum holes, and thereby increase the spectralefficiency.The fundamental task of each Cognitive radio user incognitive radio networks, in themost primitive sense, fordetection of licensed users, also called as primary users (PUs),M.SrujanaAssistant ProfessorDepartment ofElectronics andCommunication Engineering,CJITS, Yeshwanthapur,T.SE-mail:[email protected] they are present and identify the available spectrum if theyare absent.
This is usually achieved by sensing the RFenvironment, process called spectrum sensing 1–4. Theobjectives of spectrum sensing are twofold: first, CR usersshould not cause harmful interference to PUs by eitherswitching to an available band or limiting its interference withPUs at an acceptable level and, second, CR users shouldefficiently identify the spectrum holes for required throughputand quality-of service(QoS). Thus, the detection performancein spectrum sensing is very much crucial to the performanceof both primary and CR networks.In the conventional CCRN formulation, some type of resourceallocation problem was addressed, such as subchannelassignment for SUs, relay assignment, and power control 4–6. In 4, the subcarrier assignment, relay assignment, andSU relay strategy optimization problems were approachedwith flexible channel cooperation in a multi-channel CCRN,where a unified optimization framework based on NashBargaining Solutions was developed. In 5, 6, the spectrumleasing problem was formulated for one PU and multiple SUsas a Stackelberg game and the Nash equilibrium was derived.
A single channel was assumed available, and differenttransmissions were divided in time. The consideration of onechannel and one PU in 5, 6 presents a simplification forpractical scenarios where there are typically multiple channelsand multiple PUs that coexist in the coverage area of a basestation in the cellular network.A multiphase cooperation scheme is proposed in order toimprove the network utility as well as the spectrum accessopportunity. We assign the selected relaying SUs as the groupof intermediate users (IUs), which cooperate with PUs intraffic relay and share the spectrum access opportunities withthe remaining SUs, respectively. With the help of IUs, thePUs can improve their own performance as well as not beinvolved in such a complicated cooperation scheme withmultiple SUs. Meanwhile, the SUs starving for the spectrumaccess opportunities attain what they want as well.Second, an IU selection scheme is implemented by themaximum weighted bipartite matching algorithm, and theutility of the cooperating pairs is enhanced by exploiting theratio of cooperation pairs’ utility to the total energyconsumption with the consideration of the IUs’ energyefficiency.
Third, through the cooperation among the IUs andthe surrounding SUs by using cooperative network coding,the starving SUs who form a cluster can obtain thetransmission opportunities without consuming too muchenergy to relay the PUs’ traffic. Conversely, the IUs’ utilityand communication reliability can be enhanced.42. SYSTEM MODELAs demonstrated in Fig. , we consider PUs and SUs areuniformly distributed in a CCRN.
The data has beentransmitted to the BS over its own licensed channel by a basestation (BS) serves PUs and each PU, given that the spectrumsof PUs are orthogonal in frequency and/or space. access points(APs) coexist in the same area serving SUs and each SUcommunicates with its corresponding AP.The first phase cooperation is between the PU and the selectedcluster head IU, while the second phase cooperation isbetween the cluster head and other SUs in the cluster. Asshown in Fig. 2, the cooperation between SUs and Pus takesplace in a two-phase cooperation scheme in each time slot .The partner IU selection scheme is first performed, and thenthe cluster head IU cooperates with the PU in a TDMAmanner that the PU transmits its package to the cooperatingIU and the IU relays PU’s last package to the BSsimultaneously.
After the cooperation between PU and IU, theIU finds the cooperative SUs who form a cluster from thesurrounding starving SUs. Then, the IU and the SUs in thecluster cooperate by cooperative network coding.Fig. 1.
Scenario of CCRNThe channel conditions are assumed to be stable during a fixtime slot , but vary independently from one slot to another.The spectrum sharing strategy operates in a time-slottedmanner and transmission channels are assumed to conform toa Rayleigh flat fading model. The CSI is available, which isestimated by exploiting techniques such as least squares (LS)estimation and minimum mean-square- error (MMSE)estimation 9.Fig. 2. Time frame structure for the spectrum sharing strategyThe SUs, who participate in the cooperation with the PUs,send feedbacks with their transmit power values they want todevote in delivering PUs’ traffic to the BS. In order toimprove the performance of primary network, the BSbroadcasts the cooperation selection requirement to itssurrounding SUs.
If one SU can serve as the relay for multiplePUs, it sends different transmit power values corresponding toeach PU to the BS. However, in real networks, some SUsmight not be willing to cooperate with the PU, as it is quiteenergy consuming to relay PU’s traffic while the utility gainmight be relatively low, i.e., the ratio of utility to powerconsumption is low.But the SUs still desire to gain the secondary transmissionopportunities so as to improve the utility. In order to solve theaforementioned problem, the selected IU cooperates with theremaining SUs to benefit them. Meanwhile, through thecooperation between cluster head IU and other SUs in thecluster, the IU can improve its own performance as well.
As shown in Fig. 2, The time frame structure includes twocooperations: the first phase cooperation and the second phasecooperation. In the IU selection period of the first phasecooperation, after BS acquires the acknowledgement and theinformation from potential IUs, the BS exploits the maximumweighted bipartite matching algorithm to find the mostappropriate cooperative SUs, i.e., the IUs.
After partner IUselection, the PU cooperates with the IU in a TDMA manner.Then, the IU broadcasts its cooperation requirement to beginthe second phase cooperation.The SUs send the acknowledgement that they want to joininto the cooperation with the IU. After that, the IU transmitsits packet towards the associated AP. During this transmissionprocess, the surrounding SUs (form a cluster) who areinvolved in the cooperation can overhear the data. Then, byusing network coding, the SUs in a cluster create newcombinations of packets from the received packets andtransmit those towards the respective AP. The cooperationscheme among cluster head IU and SUs in the cluster isreferred as cooperative network coding, in which the IU is thesource and the corresponding AP is the destination, and theSUs form a cluster to help IU relay the data from the source tothe destination.
Energy efficiency is considered in the system by using a ratioof utility to energy, which enables a tradeoff between utilityand energy consumption. IU selection is performed to selectthe IUs who cooperate with the PUs. The IUs are a group ofSUs that have better channel conditions than other SUs torelay PUs’ traffic.3. NUMERICAL RESULTSIn this section, in comparison with the random selectionscheme, the IU selection scheme is evaluated in a CCRNsimulator.
The operation factors, e.g., cooperation timeallocation and SUs’ power consumption, are also investigated.
As shown in Fig. 1, there are 4 PUs and 6 SUs in the CCRN.The powers of PUs and SUs vary from 1mW to 2mW andfrom 0.5mW to 1.5mW, respectively.
The proposed IUselection (IS) scheme and random selection (RS) scheme, arecompared i.e. the performance obtained by using two differentschemes.5Network Utility7RSIS6.565.554.
54 6 8 10 12 14 16 18 20 2Case IndexFig. 3. Comparison of the network utility attained by two different schemesIn Fig. 4 for the BS under different values of IU’s power isevaluated, by the impact of choosing the value of .
From thecandidate SUs in the cooperation Once BS collects theinformation; BS chooses an appropriate value of and to selectthe IUs, performs the maximum weighted matching. Thewhole utility of cooperation pairs is simulated and the utilityfor different values of is demonstrated in the figure.1.81.
6alpha=4/5alpha=4.5/50.40.2 0 0.1 0.2 0.3 0.
4 0.5 0.6 0.7 0.8 0.9 1IU power mWFig.4.
Achieved utility vs. IU’s power for different valuesof .4. CONCLUSIONIn this paper, we have studied and implemented a novelcooperative spectrum sharing approach for a wireless networkconsisting of multiple primary and secondary users. we haveseen a spectrum sharing strategy based on two-phasecooperation including an IU selection scheme in CCRN. Thecooperation pairs between PUs and IUs have been obtained,By solving the maximum weighted bipartite matchingproblem. Thus we have got the maximum total utility.
Further, the energy efficiency have been considered in the IUselection problem and The selected IU cooperates with the PUas well as its surrounding SUs. With the help from the IUs thesystem utility and the spectrum access opportunity have beenimproved. With the help of simulated result we have find thatthe utility obtained by performing the proposed partner IUselection scheme is always higher than that attained by therandom selection scheme in our CCRN. In future work, wewill analyze the cooperation between the IU and thesurrounding SUs in detail.5.
REFERENCES1 Akyildiz, W. Y. Lee, M. Vuran, and S. Mohanty, “NextGeneration/ Dynamic Spectrum Access/Cognitive RadioWireless Networks: A survey,” ACM Journal onComputer Networks, vol.
50, no. 13, pp. 2127- 2159,Sep. 2006.2 A. Goldsmith, S. Jafer, I. Maric, and S.
Srinivasa,”Breaking spectrum gridlock with cognitive radios: Aninformation theoretic perspective,” in Proceedings ofIEEE, vol. 97, no. 5, pp. 894-914, May 2009.3 O. Simeone, I. Stanojev, S. Savazzi, Y.
Bar-Ness, U.Spagnolini, and R. Pickholtz, “Spectrum leasing tocooperating secondary ad hoc networks,” IEEE Journalon Selected Areas in Communications, vol. 26, no. 1, pp.
203-213, Jan. 2008.4 Y. Han, A. Pandharipande, and S.
Ting, “Cooperativedecode-and- forward relaying for secondary spectrumaccess,”IEEETransactionsonWirelessCommunications, vol. 8, no. 10, pp. 4945-4950, Oct.2009.5 N. Zhang, N.
Lu, R. Lu, J. W. Mark, and X. Shen,”Energy-efficient and trust-aware cooperation incognitive radio networks,” in Proceedings of IEEEInternational Conference on Communications, pp. 1763-1767, Jun. 2012.
66 T. Elkourdi and O. Simeone, “Spectrum Leasing viaCooperation with Multilpe Primary Users,” IEEETransactions on Vehicular Technology, vol.
61, no. 2, pp.820-825, Feb. 2012.7 V. Mahinthan, L.
Cai, J. W. Mark and, X. Shen, “PartnerSelection Based on Optimal Power Allocation inCooperative-Diversity Systems,” IEEE Transactions onVehicular Technology, vol. 57, no. 1, pp.
511-520, Jan.2008.8 M. Seyfi, S. Muhaidat, and J.
Liang, “Relay Selection inUnderlay Cognitive Radio Networks,” in Proceedings ofIEEE Wireless Communications and NetworkingConference, pp. 283-288, Apr. 2012.9 M. Biguesh and A. B. Gershman, “Training-BasedMIMO Channel Estimation: A Study of EstimatorTradeoffs and Optimal Training Signals,” IEEETransactions on Signal Processing, vol.
54, no. 3, pp.328-339,Mar. 2006.
10 “Spectrum Sharing Strategy using Bipartite Matching forCooperative Cognitive Radio Networks”Yujie Tang,Yongkang Liu, Jon W. Mark and Xuemin (Sherman)Shen Centre for Wireless Communications, University ofWaterloo, ON, Canada, N2L 3G1 Globecom