Frequency spectrum blind sensing method based on covariance matrix decomposition

Abstract

A frequency spectrum blind sensing method based on covariance matrix decomposition relates to a method used in wireless communication for secondary users to detect frequency spectrum holes. The method comprises the following steps: conducting Bartlett decomposition on a sampling covariance matrix receiving signal vector to obtain an upper triangular matrix; utilizing the quadratic sum of off-diagonal elements in the matrix and the quotient of the quadratic sum diagonal elements as statistical decision amount for detecting frequency spectrum holes; and judging the inexistence of frequency spectrum holes when the decision amount is higher than a threshold value, or else, judging the existence of frequency spectrum holes. In the sensing method disclosed by the invention, theoretical decision threshold has a simple close manner, so as to be precisely calculated out, and the theoretical decision threshold is suitable for sensing scenes having different sampling scales; and on the other hand, as a novel fully-blind sensing method, the method does not need the participation of the statistical characteristics of master user's signals and channel, as well as noise in the implementation of frequency spectrum sensing, and can effectively solve the problem of noise uncertainty encountered during the adoption of a classical energy detection method.

Description

technical field [0001] The invention relates to an implementation method adapted to idle frequency band perception in cognitive radio, and belongs to the technical field of cognitive radio in wireless communication. Background technique [0002] Spectrum is a valuable resource for wireless communications. In traditional wireless communication systems, spectrum allocation is fixed. Once a section of spectrum is authorized to a user, the user (authorized user, primary user) has exclusive rights to it, and other users (unauthorized users, secondary users) are prohibited from occupying it even if it is not used by itself. This spectrum allocation mode is simple to implement and is conducive to ensuring the quality of service of the primary user. It has been widely used at the beginning of the development of wireless communication and has been used until now. However, recent studies have shown that primary users are underutilizing licensed frequency bands while exclusively usin...

AI Technical Summary

Problems solved by technology

But a common problem is: so far, due to mathematical difficulties, the theoretical detection thresholds of the above three methods are all based on the sampling scale is infinite (for the method of maximum / minimum eigenvalue ratio and energy / minimum eigenvalue ratio method, the dimension of the sampling vector is required is also infinite), but in the actual perception scene, due to the limitation of perception time, does not It may be infinite, and the dimension of the sampling vector is generally very small

Obviously, these constraints in the actual perception scene will make the theoretical threshold obtained above extremely inaccurate, resulting in unreliable detection performance of these methods in actual perception applications, which limits their further application and promotion

[0006]The classic energy detection method only uses the energy information of the received signal to judge the existence of the primary user signal. When there is correlation in signal sampling or noise uncertainty occurs, the The detection performance of the method will drop significantly

Images