Optimal Power Allocation for Low Complexity Channel Estimation and Symbol Detection Using Superimposed Training

Qingbo WANG  Gaoqi DOU  Jun GAO  Xianwen HE  

IEICE TRANSACTIONS on Communications   Vol.E102-B   No.5   pp.1027-1036
Publication Date: 2019/05/01
Publicized: 2018/10/26
Online ISSN: 1745-1345
DOI: 10.1587/transcom.2017EBP3408
Type of Manuscript: PAPER
Category: Wireless Communication Technologies
channel estimation,  superimposed training,  computation complexity,  power allocation,  

Full Text: PDF>>
Buy this Article

A low complexity channel estimation scheme using data-dependent superimposed training (DDST) is proposed in this paper, where the pilots are inserted in more than one block, rather than the single block of the original DDST. Comparing with the original DDST (which improves the performance of channel estimation at the cost of huge computational overheads), the proposed DDST scheme improves the performance of channel estimation with only a slight increase in the consumption of computation resources. The optimal precoder is designed to minimize the data distortion caused by the rank-deficient precoding. The optimal pilots and placement are also provided to improve the performance of channel estimation. In addition, the impact of power allocation between the data and pilots on symbol detection is analyzed, the optimal power allocation scheme is derived to maximize the effective signal-to-noise ratio at the receiver. Simulation results are presented to show the computational advantage of the proposed scheme, and the advantages of the optimal pilots and power allocation scheme.