Compressed Channel Estimation Methods for High Mobility Doubly Selective Channels in Orthogonal Frequency Division Multiplexing Systems

Document Type

Article

Publication Date

10-19-2018

Publication Title

IET Communications

Volume

13

Issue

2

First page number:

205

Last page number:

215

Abstract

In this study, four channel-estimation (CE) methods are proposed for sparse high mobility doubly selective channels. These methods are designed to take advantage of the sparsity of the channel to decrease the training sequence size and the computational complexity of the CE. Two of the new methods use time domain training sequences. In the first method, the pilot is designed such that the time domain channel can be estimated tap-by-tap using an autoregressive procedure. In the second method, a truncated Taylor expansion of the channel phase is applied to linearise the equations for CE and a pseudorandom training sequence is utilised to solve those equations. The last two proposed channel estimators employ frequency domain pilots and are built upon a previously proposed estimator that exploits both the delay and Doppler shift sparsity of the channel. The computational complexity and the spectral efficiency of the methods are compared. In addition, simulations are performed to compare the performance of these estimators in a sparse channel. Simulation results indicate up to 30 dB improvement in the bit error rate calculation for our proposed CE methods compared to the conventional CE methods.

Disciplines

Electrical and Computer Engineering

Language

English

UNLV article access

Search your library

Share

COinS