An Application of the Empirical Likelihood Method to Channel Estimation with Prior Information
The empirical likelihood (EL) method combined with estimating fuctions is a hybrid of nonparametric and parametric statistical estimation methods. The purpose of this study is to examine an amount of improvement of accuracies for the channel estimation using the EL method when prior information is available. First, a simplified procedure is stated for the optimization problem to deal with the EL method with prior information. It is also shown that this procedure has the same performance with the original one. The estimation accuracies of this EL method are compared with those of the conventional metohd for estimation of a SISO channel modeled by finite response filters when the prior information is the statistics of the input signal. In the case of noisy input signal, it is seen that the EL method has better estimation accuracies.
Hideaki Sakai Hajime Yamakage
Kyoto University, Kyoto, Japan
国际会议
2011亚太信号与信息处理协会年度峰会(APSIPAASC 2011)
西安
英文
1-6
2011-10-18（万方平台首次上网日期，不代表论文的发表时间）