›› 2012, Vol. 25 ›› Issue (5): 26-.

• Articles • Previous Articles     Next Articles

Slow Time-varying Channel Estimation Based on Part-data-dependent Superimposed Training

 WANG Peng-Peng, HU Jin-Hui, HOU Hai-Tao   

  1. (School of Electric Engineering,Xidian University,Xi'an 710071,China)
  • Online:2012-05-15 Published:2012-05-24

Abstract:

This paper researches the OFDMA communication system,and proposes a new channel estimation algorithm using part-data-dependent superimposed training (PDDST).Recovery method of data at the receiver is given.Time-varying channel is modeled by complex-exponential basis expansion model (CE-BEM).The pilots of the OFDMA system have been designed carefully.In the frequency domain,the data is to be launched minus a sequence based on part of the data.After the Fourier transform,a part of data which will be sent has lower power in time domain,and errors produced by Channel estimation are greatly reduced.Since the signal launched has been distorted,we propose compensation for information data to avoid reducing the performance of data recovery.Simulations prove that this method can significantly improve the performance of channel estimation.

Key words: superimposed training;channel estimation;basis expansion model;OFDMA

CLC Number: 

  • TN914.5