Covariance matrix spectrum sensing method

Abstract

The present invention discloses a covariance matrix spectrum sensing method. The processing process of the method comprises the steps of: performing estimation in a quiet period of an authorized userto obtain noise power; in a current sensing time slot, performing multi-times sampling of signals from a plurality of channels, and forming a sample vector by employing the samples obtained through each sampling; calculating a covariance matrix according to all the sample vectors; calculating test statistics according to the sum of absolute values of values of all the elements in the covariance matrix and the noise power; and finally, determining whether the sensing time slot has authorized user signals or not through comparison of the test statistics and a judgment threshold. The covariance matrix spectrum sensing method employs the samples obtained in the sampling in the quiet period of the authorized user to estimate the noise power to accurately estimate the noise power because the quiet period does not comprise the authorized user signals; and moreover, the sum of absolute values of values of all the elements in the covariance matrix and the noise power are employed when the teststatistics is calculated to obtain a better sensing performance.

Description

technical field [0001] The invention relates to a cognitive radio spectrum sensing technology, in particular to a covariance matrix spectrum sensing method for estimating noise power during the silent period of authorized users. Background technique [0002] The rapid development of diversified mobile communication services has greatly enriched and facilitated people's work and life. However, correspondingly, the number of wireless devices and the demand for mobile data traffic have exploded, which has led to a shortage of spectrum resources. In recent years, the problem of shortage of spectrum resources has gradually emerged, and it will become more serious in the foreseeable future. However, this is not caused by insufficient physical spectrum resources, but because the existing fixed spectrum allocation strategy makes many spectrum resources unable to be fully utilized, which greatly reduces spectrum utilization. Therefore, improving spectrum utilization becomes the key ...

AI Technical Summary

Problems solved by technology

[0003] At present, the received signal covariance matrix has been widely used in the spectrum sensing of cognitive radio. Zeng et al. published "Spectrum-Sensing Algorithms for Cognitive Radio Based on Statistical Covariances" in 2009 (Spectrum Sensing Based on Statistical Covariance in Cognitive Radio The traditional covariance matrix spectrum sensing method proposed in the method) has been widely used. However, the traditional covariance matrix spectrum sensing method uses the trace of the covariance matrix of the current sensing time slot to estimate the noise power. Since the current sensing time The value of the trace of the covariance matrix of the slot is equal to the power of the received signal of the current sensing slot (the received signal may only have noise signals, or may contain noise signals and signals of authorized users), so this makes the estimated noise power and the received signal The powers are exactly the same, which in turn will cause the values ​​of the numerator and denominator of the test statistic to have the same variation trend, which will seriously degrade the performance of the spectrum sensing method

Images