Precoding based on matrix decomposition for faster-than-Nyquist signaling

The faster-than-Nyquist (FTN) signaling is a bandwidth-efficient technology which has drawn attention in the bandwidth-starved world. Since the symbols appear too early, inter-symbol interference (ISI) cannot be avoided in the receiver. Decision feedback equalization (DFE) cancellation is an efficient interference cancellation and signal detection schemes for FTN based communication system. However, since the noise isn´t considered, the DFE scheme performs badly at low E_b/N₀. In order to further improve the detection performance, firstly, the detector based on minimum mean square error (MMS E) is investigated to at low E_b/No, and then the precoding scheme is introduced in FTN signaling, including: singular value decomposition (SVD), geometric mean decomposition (GMD). Simulation results show that the GMD precoding can perform better than DFE.