基于数据协作的信道估计优化

Optimization of channel estimation based on data collaboration

为了减小大规模多输入多输出系统中导频污染对信道估计的影响,利用已知导频符号及用户发送数据独立同分布的特点,给出一种信道估计优化方案。基站接收到数据信号后,先按照行不变列扩展的方式进行矩阵扩展,再进行信道估计,如此可增加用于信道估计的已知数据长度,减小数据之间的相关性,使污染项逐渐趋于零,从而达到优化信道估计的目的。仿真结果表明,所给方案在优化信道信息上有较好的性能增益。

In order to reduce the influence from pilot frequency pollution on channel estimation in a large scale multi input multi output system,a channel estimation scheme is presented by using the fact that the known pilot symbols and the data sent by the user are independent of each other and have the same distribution.When the base station receives the data signals,before the channel estimation,the method of matrix expansion is firstly carried out on each colum in line with that each row is invariant.Thus,the length of known data for the channel estimation can be increased,the correlation between the data can be reduced,and the pollution terms can be made to zero gradually,so that,the channel estimation gets optimized.Simulation results show that,the presented scheme on the optimization of channel information can bring significant performance gain.