Research On Energy Efficiency Optimization In Cognitive Radio Networks

Future wireless networks have to face certain challenges including higher data rates,higher spectrum efficiency,lower energy consumption,and so on.Cognitive radio(CR)has been proposed as an efficient candidate in addressing the spectrum shortage challenge.It enables dynamic spectrum access(DSA)by allowing the secondary users(SUs)to access the spectrum bands allocated to the primary users(PUs)as long as they do not generate harmful interference to the PUs.However,these functions including spectrum sensing,DSA are usually considered as energy-consumption operations,which certainly increase the energy consumption and reduce the energy efficiency.Cooperative spectrum sensing(CSS)can significantly improve the sensing performance by cooperatively sharing the results from multiple cognitive radios(CR).However,CSS achieves this benefit at the sacrifice of the energy consumption and the energy efficiency reduction.Therefore,this paper mainly considers the optimization of energy efficiency,and also studies the energyand spectral-efficiency trade-off problem in cognitive radio networks.First,we focus on finding the optimal fusion rule threshold to maximize the energy efficiency in the CSS system.This problem is formulated as a fractional programming under the constraint of the false-alarm probability.By introducing a parameter ?,the EE optimization problem is reformulated into a parametric formulation problem and the closed-form formulation is derived.Then the bisection search method is proposed to search the optimal threshold with k-out-of-N fusion rule.Further,the optimization of energy efficiency problem is also presented assuming the imperfect knowledge of the channel conditions and noise variance.A cooperative spectrum sensing scheme based on the expectation maximization-bisection method algorithm(EM-BM)is proposed to jointly detect the PUs' signal and estimate the unknown channel gain and noise parameters to maximize the energy efficiency.Simulation results indicate that assuming the perfect knowledge of the CSI and noise variance,the proposed method can get the same optimal threshold with lower computational complexity compared with the exhaustive search method especially when the number of cooperative user is large.Additionally,the EM-BM scheme can iteratively attain a reliable performance,with few iterations and modest computational complexity.Compared to energy detection,higher energy efficiency can be achieved with low signal-to-noise ratio.In order to better balance the spectral efficiency and energy efficiency in CRNs,the EE-SE tradeoff issue is also investigated in this paper.Firstly,the corresponding closed-form EE functions in terms of SE have been deduced to evaluate the EE–SE tradeoff with different practical system parameters.The optimization problems are formulated as a multi-objective problem with the energy efficiency and spectral efficiency as objective functions,and the false-alarm probability and the detection probability as the constraint functions.In particular,we jointly considered each sensing times and detection thresholds as optimization variables.Different multi-objective evolutionary algorithms are employed to solve the optimization problems.Simulation results show that these algorithms are effective to find the approximation Pareto set in EE-SE trade-off issue with different scenarios.