论文部分内容阅读
目前已存在的频谱感知算法只开发了时间、频率和地理维度,而角度维频谱资源的探索,即空间频谱感知尚不成熟。在本论文中,我们将多重信号分类(Multiple Signal Classification,MUSIC)这一到达角(Angle Of Arrival,AOA)估计算法应用到频谱感知场景中。注意到MUSIC算法需要知道信源信号的数目,但是在多数感知场景中达不到这一要求。因此,我们利用特征值加权方案设计一种无需估计信源数目的加权MUSIC到达角估计算法。最后利用该算法的最大最小谱值比,我们提出了一种基于加权MUSIC算法的盲频谱感知算法。新算法保持了MUSIC算法的高峰值和高分辨率的特点,在达到较高检测概率的同时可以为频谱接入提供AOA信息,进一步提高了频谱利用率。仿真结果证明了所提算法的有效性。
The existing spectrum sensing algorithms only develop the time, frequency and geographic dimensions. However, the exploration of the angle spectrum spectrum resources, that is, the spatial spectrum sensing is not yet mature. In this thesis, we apply the Multiple Signal Classification (MUSIC), an Angle Of Arrival (AOA) estimation algorithm, to spectrum-aware scenarios. Note that the MUSIC algorithm needs to know the number of source signals, but this requirement is not met in most sensing scenarios. Therefore, we use the eigenvalue weighting scheme to design a weighted MUSIC angle of arrival estimation algorithm without estimating the number of sources. Finally, using the maximum and minimum spectral ratio of the algorithm, we propose a blind spectrum sensing algorithm based on weighted MUSIC algorithm. The new algorithm maintains the characteristics of high peak and high resolution of the MUSIC algorithm, and can provide AOA information for spectrum access while achieving a higher detection probability, thereby further improving spectrum utilization. Simulation results show the effectiveness of the proposed algorithm.