Sensor Array Based Spectrum Sensing Technology

The rapid development of wireless communication technology, like Io T, 4G, 5G, etc., will bring huge amount of wireless access and mobile data traffic. But the limited wireless spectrum resources and its allocation methods have become the key factors that restrict the development of new technology. At present, the static allocation of the spectrum has been unable to meet people's needs, and people actively seek more efficient and reasonable way of spectrum utilization. The emergence of Cognitive Radio(CR) has provided a systematic solution to the problem of dynamic and efficient use of spectrum resources, and spectrum sensing is the key technology of cognitive radio. The spectrum sensing technology is used to detect the primary users' signals actively, and find the spectrum resource which is not used effectively in the frequency domain, time domain or spatial domain, which is named spectrum hole. This provides support for the dynamic spectrum access of cognitive users, and then realizes the dynamic sharing of spectrum resources.Firstly, this paper analyzes the existing spectrum sensing algorithm, and gives the energy detection algorithm based on antenna array model, and gives the expression of its performance analysis in the antenna array model. Then, analysis the spectrum sensing method based on covariance, the core idea of this method is that the signal will react to the specific form of the covariance matrix, and the existence of the primary user's signal can be detected. On the basis of the covariance matrix spectrum sensing method, the spectrum sensing method based on the eigenvalue is emerged as the times require. The eigenvalue spectrum sensing algorithm, which is based on eigenvector subspace technique, can effectively reflect the changes of covariance. At the same time, this method can also reflect the power and the correlation of the signal, and it is a convenient and effective spectrum sensing method. In the above two kinds of spectrum sensing algorithm, the estimation accuracy of the covariance matrix will directly affect the performance of spectrum sensing. In this paper, the estimation algorithm of the covariance matrix is improved by using a two-dimensional estimation algorithm of time and space, which can effectively improve the performance of the two kinds of spectrum sensing algorithm.Because of the complex distribution of eigenvalues, it is difficult to control the key parameters for eigenvalue based spectrum sensing algorithms. For the antenna array environment, we analyze the relationship between the spatial spectral density and the eigenvalue, and explain that the spatial spectral density can be substituted for the eigenvalue. In this paper, the relationship between eigenvalues and spectral density is established by means of eigenvector subspace method, and the spectrum sensing algorithm based on spatial spectrum density is proposed. At the same time, this paper discusses the test statistic of two kinds of cases with incident angle known and unknown, and gives a unified performance analysis expression with clear physical meaning. This can effectively calculate the theoretical value of the probability of the false alarm and detection probability. Simulation results verify the validity of the method and the theoretical analysis. At the same time, the spectrum sensing algorithm has a better performance than the spectrum sensing algorithm based on eigenvalue.