线性平滑的信号调制识别方法

Signal modulation recognition based on linear smoothing

针对如何提高调制识别算法在低信噪比时的识别性能,是调制识别技术走向实用中必须解决的问题,提出采用线性平滑方法对接收信号进行预处理,然后再提取特征。提取的分类特征削弱了噪声影响,提高了抗干扰性能,能够在较低信噪比条件下正确区分信号。实验表明,平滑处理实现简单,对算法在低信噪比时的识别性能改善明显。

How to improve the performance of modulation recognition method in low signal noise ratio conditions is an important problem for its application. A method using linear smoothing to preprocess the signal is presented at first, then the key feature is extracted. With the key feature obtained, antijamming ability is improved and various signals can be recognized in low SNR. Simulation result indicates the smoothing procedure is simple to realize and the improvement is obvious.