The Local Spectrum Sensing Technique In Cognitive Radio

The rapid development of wireless communication technology cause wireless service has a huge demand for the radio spectrum resources, however, the static spectrum allocation policy has led to low utilization of spectrum resources. One solution for the spectrum shortage issue is cognitive radio, cognitive radio is an effective means to resolve the current spectrum resource constraints. Cognitive radio including spectrum sensing,dynamic spectrum access, spectrum management and spectrum sharing technology, it can automatically sense the wireless communication environment, after the adoption of the environmental decisions and learning, adaptive communication parameter access grant channel, under the premise of not causing interference to primary users, use the free spectrum for communication. In this paper, our study is the spectrum sensing technology in cognitive radio, spectrum sensing technology is the key technology that can ensure cognitive radio network can work properly. At the present time, technology of spectrum sensing contains single-node and multi-node cooperative spectrum sensing two aspects.Single-node spectrum sensing is only a secondary user detects that the presence of the primary user, multi-node cooperative spectrum sensing is a plurality of secondary users mutual cooperation detects that the presence of the primary user. The focus of this study is a single-node spectrum sensing, namely local spectrum sensing.This paper summarizes the existing local spectrum sensing algorithm. We introduces three classical spectrum sensing algorithm for the transmitter, energy detection, matched filter detection and cyclostationary feature detection algorithm, and some other algorithms, such as covariance detection, delay correlation detection and multi-step detection, and analyzes the advantages and disadvantages of each algorithm. Many existing spectrum sensing methods need to know a priori knowledge of the signal and noise, however, in reality this is difficult to achieve, difficult to meet the actual requirements of the application.Therefore, in this article we use the blind detection method, that is, without any prior information, you can determine the presence or absence of the primary user signal, it has great theoretical value and practical significance.In this paper, from the starting of the received signal in cognitive radio network, for the information loss problem in the existing spectrum sensing algorithm, according to the covariance matrix of the received signal is 0 or not, improve the existing spectrum sensing algorithm, to make it more realistic. The resulting detector significantly improved the detection performance. In addition, we get the theoretical analysis of the detection performance of the proposed method.When the complementary covariance matrix of the received signal is 0, remove the imaginary part and the real part of the received signal stitching together, structure a new received signal, according to the new structure of the received data to improve the detection performance of Hadamard. Hadamard detection method is one kind detecting device that based on generalized likelihood ratio test, it has the advantage of not require any prior knowledge and has good robustness to the non-uniform background noise. In the text, We detailed analysis the system model, established the test statistic, derived the formula of the decision threshold. Theoretical analysis and simulation results show that the scheme compared to other existing detection method has better detection performance.When complementary covariance matrix of the received signal is not equal to 0, explained the complementary covariance matrix contained part of the information of the received signal. In the text, we spliced together the transposition and conjugate transpose of the received signal, reconstructs the received signal. The covariance matrix of the newly constructed received signal, not only includes the information of the covariance matrix further comprises the information of the complementary covariance matrix of the original received signal, avoid the loss of information. We use the newly constructed received signal improvement the maximum and minimum eigenvalue ratio detection method. This method is based on the sampling covariance matrix, then eigendecomposed, used the ratio of the maximum and minimum eigenvalues as the decision statistic. The maximum and minimum eigenvalue ratio detection method has the advantage of does not require prior knowledge, does not depend on the noise power, and is simple. By theoretical analysis and simulation, prove this method superiority.