1: \begin{abstract}

2:  This paper develops a low-complexity channel estimation (CE) scheme based on

3:  compressive sensing (CS) for time-domain synchronous (TDS) orthogonal

4:  frequency-division multiplexing (OFDM) to overcome the performance loss under

5:  doubly selective fading channels. Specifically, an overlap-add method of the

6:  time-domain training sequence is first proposed to obtain the coarse estimates

7:  of the channel length, path delays and path gains of the wireless channel, by

8:  exploiting the channel's temporal correlation to improve the robustness of the

9:  coarse CE under the severe fading channel with long delay spread. We then

10:  propose the priori-information aided (PA) iterative hard threshold (IHT)

11:  algorithm, which utilizes the priori information of the acquired coarse

12:  estimate for the wireless channel and therefore is capable of obtaining an

13:  accurate channel estimate of the doubly selective fading channel. Compared with

14:  the classical IHT algorithm whose convergence requires the $l_2$ norm of the

15:  measurement matrix being less than 1, the proposed PA-IHT algorithm exploits

16:  the priori information acquired to remove such a limitation as well as to reduce

17:  the number of required iterations. Compared with the existing CS based CE method

18:  for TDS-OFDM, the proposed PA-IHT algorithm significantly reduces the

19:  computational complexity of CE as well as enhances the CE accuracy. Simulation

20:  results demonstrate that, without sacrificing spectral efficiency and changing

21:  the current TDS-OFDM signal structure, the proposed scheme performs better than

22:  the existing CE schemes for TDS-OFDM in various scenarios, especially under

23:  severely doubly selective fading channels.

24: \end{abstract}