J4 ›› 2009, Vol. 36 ›› Issue (3): 553-556.

• Original Articles • Previous Articles     Next Articles

Iterative channel estimator using superimposed training

SHI Zhe   

  1. (1. State Key Lab. of Transient Optics and Photonics, Xi'an Inst. of Optics and Precision Mechanics, Chinese Academy of Sci., Xi'an  710119, China;
    2. Graduate School of the Chinese Academy of Sciences, Beijing  100039, China)
  • Received:2008-02-28 Revised:2008-04-09 Online:2009-06-20 Published:2009-07-04
  • Contact: SHI Zhe E-mail:shizhe830412@163.com

Abstract:

Because of high bandwidth efficiency and low complexity, channel estimation based on superimposed training (ST) has received wide considerations in recent years. In this paper, a new iterative method for channel estimation is proposed based on the ST method. Using the improvement from symbol estimation after thresholding, received sequences and estimation of channel state information are used to do iterative computation in the receiver. Computer simulations show that the mean square error (MSE) and bit error rate (BER) can be reduced by an order of magnitude with the signal-to-noise ratio around 15 dB after 3~4 iterations.

Key words: channel estimation, superimposed training, multipath channels, first order statistics

CLC Number: 

  • TN911.7