【摘要】 本文提出了基于加权最小二乘支撑矢量机(WLS-SVM)学习算法的一种DCSK混沌通信系统降噪方法。给定接收信号为训练样本集,首先用最小二乘支撑矢量机(LS-SVM)对样本数据进行估计得到估计误差,根据估计误差的统计分布特性获得一个加权系数,然后再求解WLS-SVM,得到优化的接收信号的估计值,达到降噪的目的。仿真结果表明,优化后的系统误码率(BER)性能与 DCSK系统的理论噪声性能相比得到改善。更多还原

【Abstract】 This paper proposes a noise reduction method to DCSK communications system using weighted least square support vector machine (WLS-SVM). Noise reduction is achieved by first using a LS-SVM with the transmitted chaotic signal and, then associate weighting values to the error variables based upon the resulting error variables from the first step. Computer simulation of the noise performance shows that the noise reduction technique improves the overall noise performance compared with theoretical noise performance of DCSK Communications system.更多还原