平衰落信道下ML和MAP信道估计与导频符号长度的关系性能比较

Performance Compare of the Relationship between ML and MAP Channel Estimation with Pilot Symbol Length in the Flat Fading Channel

信道估计是无线通信系统必须加以解决的关键技术之一,采用导频符号辅助的方法进行信道估计是目前各类无线通信系统常用的方法。本文针对平衰落信道提出了最大似然(ML)算法和最大后验概率(MAP)估计算法,给出了ML估计和MAP估计之间的关系,仿真了MAP估计和ML估计的方差与导频符号长度的关系,提出当导频符号长度的取值超过20个符号长度时,MAP信道估计明显优于ML信道估计。

Channel estimation is one of the key technologies for wireless communication systems. Traditionally, various wireless channels are estimated by means of pilot symbols. In this paper, the maximum likelihood (ML) and maximum a posteriori (MAP) estimation algorithms are proposed to solve the estimation problem for flat fading channels. Relationship between maximum likelihood (ML) and maximum a posteriori (MAP) estimation algorithms are given. Simulations are conducted to reveal the relationship between the error covariance and the length of pilot symbols for ML and MAP algorithm respectively. Simulation results demonstrate that when the length of pilot signal is longer than 20 symbols, Performance of MAP channel estimation algorithm is better than ML algorithm.