RANDOM INTERRUPTIONS IN COOPERATION FOR SPECTRUMSENSING IN COGNITIVE RADIO NETWORKS
ABSTRACT:
In this paper, a new cooperation structure forspectrum sensing in cognitive radio networks is proposed whichoutperforms the existing commonly-used ones in terms of energyefficiency. The efficiency is achieved in the proposed design byintroducing random interruptions in the cooperation processbetween the sensing nodes and the fusion center, along witha compensation process at the fusion center. Regarding thehypothesis testing problem concerned, first, the proposed systembehavior is thoroughly analyzed and its associated likelihood-ratiotest (LRT) is provided. Next, based on a general linear fusion rule,statistics of the global test summary are derived and the sensingquality is characterized in terms of the probability of false alarmand probability of detection. Then, optimization of the overalldetection performance is formulated according to the Neyman-Pearson criterion (NPC) and it is discussed that the optimizationrequired is indeed a decision-making process with uncertaintywhich incurs prohibitive computational complexity. The NPC isthen modified to achieve a good affordable solution by usingsemidefinite programming (SDP) techniques and it is shown thatthis new solution is nearly optimal according to the deflectioncriterion. Finally, the effectiveness of the proposed architectureand its associated SDP are demonstrated by simulation results.
EXISTING SYSTEM:
Modeling and performance optimization of cooperativesensing schemes has long been of great interest. Thelikelihood-ratio test (LRT) is known as the optimalfusion method when the distributed nodes report their sensingoutcomes to the FC through nonideal analogue communicationlinks. The effect of reporting channel impairments onthe overall sensing performance has been investigated in. Particularly, the authors in compare the performancesof the hard-decision- and soft-decision-based fusion methodsand illustrate that, in general, the soft decision significantlyoutperforms the hard decision when nonideal reporting channelsare considered. Although LRT is commonly considered asthe best soft-decision-based cooperation, finding the optimalLRT thresholds for individual nodes and for the FC incursprohibitive computational complexity. Consequently,optimal linear combining has been suggestedas a very good alternative which provides nearly optimalresults at affordable computational cost. As a fast alternativelinear combining approach with a very good performance,the so-called deflection criterion,a.k.a., deflectioncoefficient (DC) or its modified version, modified deflectioncoefficient (MDC) is commonly used inthe literature to design linear fusion schemes in cooperativesensing scenarios.
PROPOSED SYSTEM:
we extend the works in andpropose a novel energy-efficient cooperative sensing schemefor CRNs. The efficiency is achieved by adding two new mechanismsto the commonly-used cooperative sensing structure.The first mechanism is realized as a set of random energysavinginterruptions in the cooperation between the sensing CRnodes and the FC, while the second mechanism is a compensationprocess at the FC. This compensation, which is realized asa linear estimator, aims at recovering the local test summariesout of degradations caused jointly by the interruptions andreporting channel contaminations. The estimation of the localtest summaries is realized in the proposed system by using thespatio-temporal cross-correlations of the sensor outcomes aswell as autocovariance functions characterizing the behavior ofthe reporting channels. The proposed cooperation method canalso be used to increase the efficiency of the multiband jointdetection scheme in where the optimal linear combiningis used as the fusion methodWe model and thoroughly analyze the proposed system toderive the global test summary statistics in terms of probabilitydistributions based on which the controllers work. By usingthe statistics obtained, we formulate the proposed systemperformance optimization according to two commonly-usedcriteria, namely the Neyman-Pearson criterion (NPC) and DC.Specifically, first, we analyze the global test summary statisticsand characterize the system performance by providing closedformrelations for the probability of false alarm and theprobability of detection. We then derive the detection thresholdfor a fixed false alarm probability and formulate the systemperformance optimization based on NPC and subject to aconstraint on the energy consumed at the local sensing andreporting phases.
CONCLUSION
In this paper, a novel energy-efficient structure for spectrumsensing in CRNs has been proposed based on making randominterruptions in the cooperation process a the CRs. Theproposed system has been thoroughly modeled and analyzed,and an optimization problem has been developed in orderto formulate a tradeoff taking into account the energy consumptionat the local sensing and reporting processes jointlywith the overall detection performance. Analytical solution ofthe optimization problem and the presented numerical resultsdemonstrate that, significant levels of energy efficiency can beachieved by the proposed architecture. This energy efficiencyis due to the fact that, unlike in existing cooperative sensingschemes, in the proposed design the discrimination betweenreliable and unreliable nodes is obtained while no energy iswasted. Moreover, the sensitivity of the overall detection todegradations in the local sensing and reporting processes issignificantly reduced by the proposed architecture, leading toa more reliable cooperative spectrum sensing.
REFERENCES
[1] J. Mitola, “Cognitive radio—an integrated agent architecture for softwaredefined radio,” Ph.D. dissertation, Royal Inst. Technol. (KTH),2000.
[2] I. F. Akyildiz, B. F. Lo, and R. Balakrishnan, “Cooperative spectrumsensing in cognitive radio networks: A survey,” Physical Commun.,vol. 4, no. 1, pp. 40–62, March 2011.
[3] H. V. Poor, An introduction to signal detection and estimation. BerlinHeidelberg: Springer-Verlag, 1994.
[4] B. Chen and P. K.Willett, “On the optimality of the likelihood-ratio testfor local sensor decision rules in the presence of nonideal channels,”IEEE Trans. Inf. Theory, vol. 51, pp. 693–699, Feb. 2005.
[5] Z. Quan, S. Cui, and A. H. Sayed, “Optimal linear cooperation forspectrum sensing in cognitive radio networks,” IEEE J. Sel. Topics SignalProcess., vol. 2, no. 1, pp. 28–40, Feb. 2008.
[6] S. Chaudhari, J. Lunden, V. Koivunen, and H. V. Poor, “Cooperativesensing with imperfect reporting channels: Hard decisions or softdecisions?” IEEE Trans. Signal Process., vol. 60, no. 1, pp. 18–28, Jan.2012.
[7] S. Chaudhari, J. Lundén, V. Koivunen, and H. V. Poor, “BEP walls forcooperative sensing in cognitive radios using K-out-of-N fusion rules,”Signal Process., vol. 93, no. 7, pp. 1900–1908, July 2013.
[8] S. Chaudhari and V. K. J. Lunden, “BEP walls for collaborativespectrum sensing,” in Proc. IEEE Int. Conf. Acoustics, Speech and SignalProcessing (ICASSP), May 22-27 2011, pp. 2984–2987.
[9] S. Chaudhari, , and V. Koivunen, “Effect of quantization and channelerrors on collaborative spectrum sensing,” in Proc. 43rd Asilomar Conf.Signals, Syst., and Comput., Nov. 2009, pp. 528–533.
[10] G. Taricco, “Optimization of linear cooperative spectrum sensing forcognitive radio networks,” IEEE J. Sel. Topics Signal Process., vol. 5,no. 1, pp. 77–86, Feb. 2011.
[11] Z. Quan, W. Ma, S. Cui, and A. H. Sayed, “Optimal linear fusion fordistributed detection via semidefinite programming,” IEEE Trans. SignalProcess., vol. 58, no. 4, pp. 2431–2436, 2010.