Positioning Error Impact Compensation through Data-Driven Optimization in User-Centric Networks

TL;DR

The paper tackles the sensitivity of user-centric ultra-dense networks to UE/DBS positioning errors that distort Szone-based scheduling and associations. It introduces a data-driven optimization and error compensation (DD-OEC) framework that uses AutoML to model the COP–KPI relationship on erroneous data (Model-E) and a residual-learning model (Model-R) trained on historical error-free data to correct for residual errors, enabling a joint optimization of DBS density $\lambda_{dbs}$, Szone radius $R_{sz}$, and transmit power $P_{tx}$ to maximize both $\theta_{se}$ and $\eta_{ee}$ via a weighted objective $f_{obj}$. The approach combines synthetic data generation (UC-SyntheticNET), AutoML regression, and meta-heuristic optimization (GA and SA) to derive COPs that outperform a baseline by up to 23% in simulations. This framework offers a practical means to counteract localization inaccuracies in AI-driven wireless networks, improving ASE and EE in UCUDNs while relying on historical data and current erroneous measurements for robust optimization.

Abstract

The performance of user-centric ultra-dense networks (UCUDNs) hinges on the Service zone (Szone) radius, which is an elastic parameter that balances the area spectral efficiency (ASE) and energy efficiency (EE) of the network. Accurately determining the Szone radius requires the precise location of the user equipment (UE) and data base stations (DBSs). Even a slight error in reported positions of DBSs or UE will lead to an incorrect determination of Szone radius and UE-DBS pairing, leading to degradation of the UE-DBS communication link. To compensate for the positioning error impact and improve the ASE and EE of the UCUDN, this work proposes a data-driven optimization and error compensation (DD-OEC) framework. The framework comprises an additional machine learning model that assesses the impact of residual errors and regulates the erroneous datadriven optimization to output Szone radius, transmit power, and DBS density values which improve network ASE and EE. The performance of the framework is compared to a baseline scheme, which does not employ the residual, and results demonstrate that the DD-OEC framework outperforms the baseline, achieving up to a 23% improvement in performance.

Figures (2)

• Figure 1: Positioning error modeling, and the DD-OEC framework for the data-driven optimization and error compensation COP-KPI optimization. This framework includes synthetic data generation, AutoML based model fitting, and multi objective optimization.
• Figure 2: Performance of the DD-OEC framework with the optimization techniques of SA and GA for $\alpha_{se}=0.5$.