Research On Collaborative Spectrum Sensing Technology In Cognitive Radio Networks

Spectrum sensing technology is the core technology of realizing cognitive radio technology and its application and building cognitive radio networks. And also, it is an important prerequisite for protecting licensed users from harmful interference and improving spectrum utilization of cognitive radio networks. This paper, according to detection requirements of rapidity, accuracy and highly sensitivity, made an intensive research of the single and twice collaborative spectrum sensing technology.All subjects discussed in this paper are on the basis of Ganesan's cooperative spectrum sensing scheme, considered cognitive users transmission power constrained, and the relative positions between the two cognitive users and the licensed user in the three-dimensional space randomly distributed, and suffering from context effects of geographical information, the RF environment and the environmental background. Based on these conditions, a new model to estimate the performance advantages and disadvantages of collaborative spectrum sensing was proposed, and a number of relevant propositions and optimization programs from different dimensions were given to improve the speed and accuracy of sensing. The feasibility of the model is well explained and supported by extensive simulations. The study results indicate that collaborative spectrum sensing performance are affected by path loss exponent, fading factor, the relative position changes between cognitive users and primary user and the changes of transmission power, but optimization can be achieved under certain conditions, even to great extent be improved. In addition, the study shows that these optimized programs also adapt to multi-user cognitive radio networks.This paper also discussed, in a two cognitive radio user network, a proposal to estimate the cooperative sensing performance by the detected probability of correct sensing for the absence of the licensed user and put forward an optimal target function, the numerical simulations and Monte Carlo simulations show the sensing accuracy can be improved by cooperation. To multi-user cognitive radio networks, this paper focused on a twice collaborative spectrum sensing model combined of distributed and centralized. Numerical simulations and Monte Carlo simulations qualified the accuracy improvement when utilizing the reasonable decision fusion rule at the second collaborative process. In a cognitive radio network with a large number of CR users, considered the control of the signal overhead, system complexity and other factors, the joint optimal rules on the number of CR users and the voting value were proposed under the condition of ensuring the minimum average error rate and verified by theoretical analyses and Monte Carlo simulations.