基于循环频率特征的超宽带通信信号识别仿真

现有的使用较为广泛的三种超宽带通信信号识别方法均具有着频偏幅值大、定时误差大的缺陷。为了解决上述问题,提出基于循环频率特征的超宽带通信信号识别方法。根据信号识别的需求,对超宽带通信信号识别方法进行整体框架设计。采用快速循环估计算法计算超宽带通信信号循环谱,并利用映射方法将其转换为信号图域,通过索引算法对信号图域特征进行提取与分离,得到信号图域特征集合,以此为依据采用聚类分类方法对信号图域进行分类,通过四阶累积量识别方法对信号类别进行判定,实现超宽带通信信号的识别。通过仿真得到,与现有的三种超宽带通信信号识别方法相比较,提出的超宽带通信信号识别方法极大的降低了频偏幅值与定时误差,充分说明提出的超宽带通信信号识别方法具备更好的识别性能。

Ultra-Wideband Communication Signal Recognition Simulation Based on Cyclic Frequency Characteristics

In order to solve the problems of large amplitude offset and large timing error in traditional methods, a method to identify ultra-wideband communication signal based on cyclic frequency feature was proposed. According to the demand of signal recognition, the overall framework of recognition method of ultra-wideband communication signal was designed. First of all, the fast cyclic estimation algorithm was used to calculate the cyclic spectrum of ultra-wideband communication signal, and the mapping method was used to convert the cyclic spectrum into signal domain. Then, signal domain features were extracted and separated by indexing algorithm, and the feature set of signal graph domain was obtained. On this basis, the clustering classification method was used to classify the signal domain. Finally, the four-order cumulant recognition method was used to judge the signal category, so that the recognition for the ultra-wideband communication signal was realized. According to simulation results, we can see that compared with the existing ultra-wideband communication signal recognition methods, the proposed method greatly reduces the frequency offset amplitude and timing error. Certainly, the proposed method has better recognition performance.