Cholesky-decomposition Aided Linear Precoding and Decoding for FTN signaling

Due to the higher sampling rate, faster-than-Nyquist (FTN) signaling has been considered as a promising technique to improve the spectral efficiency. However, the high sampling rates also introduce severe inter-symbol-interference (ISI). Therefore, how to handle the incurred ISI is of the great significance to realize its potential benefits. In this paper, we propose a Cholesky-decomposition aided linear precoding and decoding scheme for FTN signaling. The proposed scheme is proved to fully cancel the incurred ISI, which renders the optimal bit error rate for any sampling rate theoretically. Numerical results demonstrate that the proposed scheme has lower computational complexity and decoding matrix storage space consumption, although it is workable for a relative limited range of acceleration factor in comparison with the singular value decomposition based one in practice.