On the Impact of Precoding Errors on Ultra-Reliable Communications

Abstract

Motivated by the stringent reliability required by some of the future cellular use cases, we study the impact of precoding errors on the SINR outage performance for various spatial diversity techniques. The performance evaluation is carried out via system-level simulations, including the effects of multi-user and multi-cell interference, and following the 3GPP-defined simulation assumptions for a traditional macro case. It is shown that, except for feedback error probabilities larger than 1%, closed-loop microscopic diversity schemes are generally preferred over open-loop techniques as a way to achieve the SINR outage performance required for ultra-reliable communications. Macroscopic diversity, where multiple cells jointly serve the UE, provides additional robustness against precoding errors. For example, a 4$\,\times \,$4 MIMO scheme with two orders of macroscopic diversity can achieve the 0 dB SINR outage target at the $10^ {-5}$-th percentile, even for a precoding error probability of 1%. Based on the obtained results, it is discussed what transmission modes are more relevant depending on the feedback error constraint.