Sparse Vector Coding-based Superimposed Transmission for Short Packet URLLC

Abstract

Sparse vector coding (SVC) based on position indices is emerging as a key enabler for short packet ultra-reliable and low latency communications (URLLC), since it can achieve better block error rate (BLER) performance and lower transmission latency. In this paper, we propose a SVC-based superimposed transmission (SVC-ST) scheme to further enhance the transmission reliability of SVC. At the encoding side, a portion of transmission bits is represented by non-zero position indices of sparse vector. The remaining bits are equally split into multiple streams and then mapped into the related positions of sparse vector via quadrature amplitude modulation (QAM) with constellation rotation. After random spreading, a transmission vector superimposed by multiple QAM signals is formed on frequency resource blocks. Different from SVC scheme, the non-zero position indices and related elements of sparse vector in SVC-ST scheme are respectively represented different bit information. For the SVC-ST decoding, multipath matching pursuit (MMP)-based soft decoding (MMP-SD) scheme is adopted to recover the transmission packet. Simulation results indicate that the proposed SVC-ST scheme outperforms existing SVC schemes in terms of BLER performance.