| 摘要: |
| 在小样本、低信噪比条件下，同步参数估计会存在较大误差，从而导致信号解调性能的降低。为了解决该问题，将信号解调看成有限长度采样样本的学习问题，并利用支持向量机（SVM）良好的学习性能，在存在同步误差的条件下，通过提高判决端的处理能力来改善系统的接收性能，提出了基于SVM的信号解调算法。在Matlab环境下，对提出算法在精准同步、残存同步误差、高斯白噪声和高斯色噪声等情况进行了计算机仿真，结果表明，相比于匹配滤波器算法，基于SVM的信号解调算法能较好地克服定时误差和相位误差以及色噪声对解调性能的影响。 |
| 关键词: 信号解调 支持向量机 相位误差 定时误差 |
| DOI： |
|
| 基金项目:国家自然科学基金资助项目（61403421） |
|
| A signal demodulation algorithm based on SVM |
| TENG Xiaoyun,XU Jun,CHEN Deming |
| () |
| Abstract: |
| The estimation error of synchronization parameters will worsen the demodulation performance,when the training data is limited and the signal-to-noise ratio(SNR) is relatively low.For this problem,by treating demodulation as a learning problem using limited data which can be solved efficiently by support vector machine(SVM),a signal demodulation algorithm based on SVM is presented，which can improve the receiving performance through reforming the classification ability. The simulations are carried out with Matlab to compare the receiving performance in different circumstances such as precise synchronization,synchronization error,white Gaussian noise,colored Gaussian noise. The results show that the proposed algorithm can efficiently overcome the influence of synchronization error and colored noise on demodulation performance. |
| Key words: signal demodulation support vector machine phase error timing error |
