Channel Estimation for Amplify-and-Forward Relay Networks with Both Time and Frequency Offsets

Cooperative multi-relay networks are asynchronous both in time and frequency inherently due to their spatial distributed nature. This paper addresses channel estimation for amplify-and-forward relay networks using training sequences in the presence of both time and frequency offsets. A practical training scheme is provided to reduce the complexity at the relay nodes and to combat the effects of time and frequency offsets at the destination. By exploiting the training scheme, a piecewise channel estimator with respect to signal-to-noise ratio (SNR) is proposed to reduce the computational complexity of the minimum mean square error (MMSE) method. Simulation results show that our method approaches the MMSE estimator and outperforms the linear least-square (LS) estimator.