Channel Map-Based Statistical CSI Acquisition and Performance Analysis for HRLLC Systems

Abstract

Hyper-reliable and low-latency communication (HRLLC) constitutes a pivotal application scenario for sixth-generation (6G) wireless communication systems, with its implementation contingent upon the quality of acquired channel state information (CSI). The channel map has emerged as a promising paradigm to enhance CSI accuracy by exploiting environmental geometry. In this paper, a channel map-based statistical CSI acquisition method is proposed, which leverages multipath information in environment to improve the CSI acquisition accuracy. Then, a general performance analysis framework applicable to HRLLC multiple-input multiple-output (MIMO) systems is established, jointly considering the pilot contamination, channel spatial correlation, channel aging, and channel map accuracy. Closed-form expressions for key performance metrics are derived, including normalized mean square error (NMSE) of minimum mean square error (MMSE) channel estimator and lower bounds on signal-to-interference-plus-noise ratio (SINR) with maximum ratio (MR) schemes. Based on that, significant effects of environmental parameters and channel statistical properties on HRLLC systems are revealed and further illustrated by simulations. Subsequently, the achievable data rate (ADR) of HRLLC systems utilizing different error-level channel maps are compared and analyzed, demonstrating performance enhancement from channel maps even when employing imperfect channel maps. The fundamental trade-off relationship between ADR and reliability is investigated, which is further improved through the block length optimization.