类Haar小波与数字信号调制识别

Modulation Identification of Digital Signals by Constructing Quasi-Haar Wavelet

调制识别在通信侦察和通信对抗中有着重要应用。为了利用小波变换进行数字调制的类型识别,提出了具有比Haar小波更好频率局域化特征的类Haar小波概念,从理论上论证了正弦型类Haar小波用于MPSK和MFSK信号调制识别的可行性,详细推导了小波系数幅度与相位跳变或频率跳变之间的关系,并对分类识别器进行了仿真。理论和仿真结果均表明,在一定信噪比条件下,正弦型类Haar小波不仅能用于通信信号的调制识别,且具有比Haar小波更好的抗噪性能。

Modulation identification has important applications in communication surveillance and countermeasure. In order to identify the digital modulation types by the wavelet transform, the Quasi-Haar wavelet is put forward as a choice which has better localization characteristics in frequency domain than that of Haar wavelet. The feasibility of using the sinusoidal Quasi-Haar wavelet to classify MPSK and MFSK is theoretically analyzed. The relationship between the amplitude of wavelet coefficients and the frequency or phase changes is deduced, and the simulation is made to the modulation identifier based on the proposed Quasi-Haar wavelet. Both the theoretical analysis and the simulation results show that the sinusoidal quasi-Haar wavelet not only can be used in modulation identification of communication signals, but also has better anti-noise performance under the appropriate SNR.