对流层散射下MPLS重路由信号去噪研究

针对对流层散射信道MPLS重路由信号数据含有噪声的问题,分析了Facility重路由原理和MPLS标签转发过程,引入小波阈值去噪算法对含噪声标签信号数据进行处理。利用一种随分解层数变化的阈值和折中阈值函数规则,来解决传统算法中的阈值确定和阈值函数选取存在的问题,针对折中阈值函数规则中参数a的确定问题,利用模拟数据,以信噪比(SNR)为目标函数,采用定步长变a形式,从而得到去噪效果最好时对应的a值,并用实测数据加以验证。仿真结果表明,改进后的算法在信噪比上有1.5 d B~3.4 d B的提高,均方差(RMSE)下降约0.01~0.03,为路由修复信号的去噪提供一种可行的方案。

Research on the De-noising of Reroute Signals in the Troposphere Scattering with MPLS

In order to remove noises from the reroute signals with MPLS,which comes from troposphere scattering,the theory of Facility reroute and the process of MPLS is analyzed,as well as the wavelet threshold denoising Arithmetic is presented. In the traditional method,the way how to choose threshold and threshold function has shortcomings,the threshold changing with the decomposition orders and the middle course threshold function can availably solve this problem. To gain the value of parameter a,which has influence on threshold function,the parameter changed with constant step length,and the SNR is the cost function. The result is verified by the measured data. Simulated and experimental signals had been used to test the effect of this new arithmetic. The results show that the SNR increases by 1.5~3.4 dB and the RMSE reduces about 0.01~0.03. This arithmetic provides a new way for the reroute signals de-noising.