Optimal Precoding Based Spectrum Compression for Faster-Than-Nyquist Signaling

EasyChair Preprint no. 190

Abstract

Faster-than-Nyquist (FTN) signaling is capable of improving the spectral efficiency by offering a higher information rate, while preserving the signaling bandwidth. In this paper, preceding the FTN modulation, a precoding based data spreading is utilized to introduce an artificial interference, which in the frequency domain shapes the signal spectrum and compresses the transmission bandwidth. In this scheme, the spectral efficiency is improved in both time and frequency domains. Further, we optimize the precoder by maximizing the ultimate system capacity and by maximizing the minimum Euclidean distance between the modulated symbols. The spectrum mask limitations are also considered for an imposed constraint on the optimization. Simulation results demonstrate that the 16-ary quadrature amplitude modulation (16-QAM) signaling can achieve the same spectral efficiency as the 64, 256-QAM Nyquist signaling, while the signal-to-noise ratio gains are about 2.5 dB and 5 dB, respectively. Furthermore, the proposed scheme outperforms the existing FTN system in terms of energy performance, noise immunity and boosts the achievable capacity limit of the system subject to the mask.

Keyphrases: Capacity Analysis, Faster-than-Nyquist(FTN), partial response signaling (PRS), precoded ftn signaling, shaping pulse design, spectrum mask

@Booklet{EasyChair:190,
  author = {Wen Shan and Liu Guanghui and Chen Qiang and Qu Huiyang and Wang Yanyan},
  title = {Optimal Precoding Based Spectrum Compression for Faster-Than-Nyquist Signaling},
  howpublished = {EasyChair Preprint no. 190},
  doi = {10.29007/dpx4},
  year = {EasyChair, 2018}}