摘要
提出了一种新的多进制频移键控(MFSK)信号的调制分类及参数估计方法。首先利用多频带滤波器组对MFSK信号进行预处理,以减弱噪声对信号的影响;然后对其中的最大频带输出值应用能量算子来提取其分类特征;并采用无监督聚类算法求取了最佳聚类数M,完成了MFSK信号的分类。同时还利用能量算子来提取信号码元跳变处的能量变化信息,从而估计出MFSK信号的符号率。该方法计算简单,抗噪性能好,具有较高的精度。计算机仿真结果证实了所提方法的有效性和正确性。
A classification and parameter estimation algorithm for MFSK signal is presented. Firstly, the method filters FSK signal through a bank of passband filters to reduce the noise effect. Then, the energy operators are applied to the largest filter output to obtain the classified feature. Finally, the unsupervised clustering algorithm is used to obtain the optimal cluster number. Meanwhile, the energy operators are also used to extract the energy transient information to estimate the symbol rate. The approach is robust against the noise and simple in computation. Simulation results prove that the method performs well.
出处
《数据采集与处理》
CSCD
北大核心
2010年第4期479-483,共5页
Journal of Data Acquisition and Processing
关键词
能量算子
聚类算法
识别
参数估计
energy operator
clustering algorithm
classification
parameter estimation
