基于时频点集特征提取方法的宽带频谱感知

Wideband Spectrum Sensing Based on Time-Frequency Point Set Feature Extraction

提出了一种基于调制宽带转换系统(MWC)和信号时频点集特征提取的宽带频谱感知方法。首先,利用MWC实现宽带模拟信号的低速率数字采样,对采样序列进行频谱重构得到各子带信号,同时宽带信道被划分为互不重叠的子带;其次,对各子带信号进行时频聚集性分析,提取时频点集特征,完成各子带信号检测,定义宽带感知向量以确定宽带频谱的具体占用情况,实现宽带频谱感知;最后进行仿真实验,结果表明了时频点集特征提取方法的可行性,以及其对噪声的不敏感特性,验证了该宽带频谱感知方法的有效性。

The paper proposes a wideband spectrum sensing method based on modulated wideband converter（MWC）system and time-frequency point set feature extraction.First,the MWC system is used to implement wideband analog signal sampling at a low rate, the original frequency spectrum is restored to obtain digital signals and the whole wideband spectrum is divided into non-overlapping sub-bands.Then,the signal of each sub-band is detected by applying time-frequency gathering density analysis algorithm and extracting time-frequency point set features. The specific wideband frequency spectrum occupation is determined by defining the wideband spectrum sensing vector,and the wideband spectrum sensing is realized at the same time. Finally,the simulations experiment results show the validity of the method by indicating that the method of extracting time-frequency point set features is feasible and it is not sensitive to noise uncertainty.