基于粒子滤波器的盲辨识和盲均衡新方法

Novel blind identification and equalization new method based on particle filter

提出了一种新的基于粒子滤波器的盲辨识及盲均衡算法。在对信道进行辨识时,通过对信道均值的采样来代替对真实信道的采样,避免了对信道的后验密度进行采样,从而降低了算法的复杂度。算法还采用先验密度作为重要性函数,以便于对重要性函数进行采样。仿真结果表明,该算法收敛速度快,所需的数据量少,在信噪比较低时也能完成对信道的盲辨识和盲均衡。

Based on particle filter theory, a new novel blind identification and equalization algorithm was proposed sampling of the posterior probability distribution was avoided by sampling the expectation instead of actual values of the channel, so as to reduce the complexity of algorithm. In addition, the importance sampling process could be simplified greatly by choosing the prior density as the importance function. Simulation results show that the algorithm has fast convergence performance and be effective even under the circumstances of very low signal-to-noise ratio and less observation data.