| With the rapid development of wireless communication business,the available spectrum resource has became more and more scarcity. However, on the other hand, a large part of available spectrum resource increasing scarce, it makes a low spectrum utilization. So it’s urgent to develop a new technique. As the precondition the Cognitive Radio (CR) technology is emerged.This paper studies the spectrum sensing technology of cognitive radio, discusses the theoretical model’s advantages and disadvantages of the transmitter and receiver in detail.The detection principles of matched filter detection, energy detection, cyclostationary feature detection are analyzed in detail. Aiming at the shortcomings of single user detection, The paper draw forth the multi-user collaboration spectrum detection technology. The paper discusses the hard detection fusion and the soft detection fusion’s advantage and disadvantage.The corresponding analysis of detection performance is shown.Traditional cooperative spectrum sensing which is based on "K-Rank" rule can’t ensure the primary user service quality, the detection performance will decline if there exists the malicious user,so we proposed the improved "K-Rank" role,and demonstrate the algorithm’s superiority through the simulation experiments. For a better understanding of the weighted,we introduce a kind of weighted method based on SNR in cooperative spectrum sensing method. In order to avoid the channel resource’s waste and congestion, improve the performance of cooperative spectrum sensing, the paper proposed an improved cooperative spectrum sensing based on the traditional cooperative spectrum sensing. We removing the poor performance nodes and choose some CR nodes with good signal to noise ratio performance to detect. Increase the sensing times when afford the sensing time will obtain a better performance. The simulation result demonstrate the effectiveness and feasibility of the algorithm, also lay the foundation for further study.Finally in order to overcome the influence to the spectrum sensing from the noise power’s uncertainty, this paper proposes non-uniform distribution of spectrum sensing based on entropy index. According to the exponential entropy’s difference between the primary user’s exist and absent, the value of exponential entropy of received signal is estimated and then compared it with the threshold to judge whether the primary user’s signal exist or not.This method doesn’t need the primary user’s prior knowledge and can resistant the noise power’s uncertainly, also get better performance at a low signal to noise ratio environment. |
