Throughput Maximization of Low-Latency Communication with imperfect CSI in Finite Blocklength Regime

Authors

Y. Zhu, Y. Hu, Z. Chang, A. Schmeink,

Abstract

        We consider a low-latency communication network operating with finite blocklength (FBL) codes. During the transmission, the minimum mean squared error (MMSE) channel estimation is assumed to be applied to obtain the instantaneous but imperfect Channel State Information (CSI) for the rate selection. We aim at optimizing the FBL throughput of the system under given reliability constraints. First, we provide an optimal frame structure
design by optimally allocating the total frame length for MMSE training of channel estimation and data transmission. In addition, we further improve the FBL throughput considering channel dynamics, which optimally selects the coding rate per frame. Combining the frame structure and the coding rate selection, a joint optimization problem is studied and solved by a sub-optimal algorithm. In the simulation study, we validate the proposed analytical model and evaluate the FBL throughput of the proposed solution in comparison to benchmark schemes.

BibTEX Reference Entry 

@inproceedings{ZhHuChSc0,
	author = {Yao Zhu and Yulin Hu and Zheng Chang and Anke Schmeink},
	title = "Throughput Maximization of {Low-Latency} Communication with imperfect {CSI} in Finite Blocklength Regime",
	booktitle = "{IEEE} Wireless Communications and Networking Conference 2019 (WCNC 2019)",
	address = {Marrakech, Morocco},
	note = {preliminary metadata entry, documents might follow soon — stay tuned!}
	}

Downloads

 Download bibtex-file

Sorry, this paper is currently not available for download.