Skip to content
Preprint

Multi-Agent Reinforcement Learning for Base Station Placement in TDOA-Based Localization

Jul 2026 · 0 citations · 16 references
Engineering

TL;DR

Findings indicate that incorporating site-specific propagation data into the placement process can match and selectively improve upon purely geometric strategies, motivating further work toward consistent gains.

Abstract

Accurate localization of devices is a key capability for emerging 5G and 6G networks and depends on effective base station (BS) placement. Conventional geometry-based approaches such as Geometric Dilution of Precision (GDOP) ignore realistic propagation effects such as Non-Line of Sight (NLOS) shadowing and multipath-induced Time of Arrival (TOA) bias caused by buildings. This paper proposes a ray-tracing-assisted Multi-Agent Reinforcement Learning (MARL) framework for environment-aware BS placement in Time Difference of Arrival (TDOA) localization systems. Proximal Policy Optimization (PPO) agents are trained on Channel Impulse Responses (CIRs) generated from a detailed 3D model of a university campus. Each agent cooperatively places one BS while optimizing a shared reward that combines localization accuracy and coverage. The approach is evaluated on five campus segments with varying propagation characteristics. Results show that the learned policy achieves localization accuracy comparable to conventional GDOP-based placement, lowering the average localization Mean Absolute Error (MAE) by about 3 % relative to the stronger (mean-optimized) geometric baseline. The behavior is segment-dependent, with a clear improvement on individual segments (up to about 14 %) and comparable or slightly higher error on the others. These findings indicate that incorporating site-specific propagation data into the placement process can match and selectively improve upon purely geometric strategies, motivating further work toward consistent gains.

View source

Similar papers

Open access Sep 2026

Optimal base station selection for GNSS-denied UAV localization

Accurate and robust localization of UAV swarms in GNSS-denied urban environments remains a major challenge because signal blockage, multipath, and poor anchor geometry can severely degrade positioning performance. This paper extends our previous work on degree-centrality-based clustering for UAV swarm relative locali...

Asma Boulkaid, Aurelio Bermúdez, Rafael Casado et al. · 0 citations
2026

Uncertainty-Aware GP Bayesian Optimization for UAV-Based Closed-Loop RSSD Interference Localization

Unmanned aerial vehicle (UAV)-based interference source localization is increasingly important in dense 5G-Advanced and emerging 6G networks, where unknown interferers degrade communication reliability and bias measurement-driven channel modeling. However, propagation uncertainty caused by multipath, blockage, and non-...

Guo-Chen Gu, Zhi-Peng Lin, Yi-Ran Chen et al. · 0 citations
Conference Sep 2026

Localization method integrating single-station long-baseline PDOA and multi-station TDOA

Passive localization plays an important role in electronic warfare and target surveillance. Conventional single-measurement approaches based on Time Difference of Arrival (TDOA) or Phase Difference of Arrival (PDOA) often suffer from limited positioning accuracy and poor robustness. To address the phase ambiguity probl...

Lin Liu, Ce Liang, Jun-Shuai Yang et al. · 0 citations
2026

Optimal Sensor Placement Using Decentralized TDOA Sensor Network in Source Localization

Most existing studies for time difference of arrival (TDOA)-based source localization adopt a centralized framework that utilizes one sensor as a reference to generate arrival time differences. This work considers adecentralized sensor network to acquire TDOAs for localization, which has lower communication cost, less...

K. Tang, Sheng Xu, Xian-Liang Li et al. · 0 citations
2026

A Proximal Gauss–Newton Method for Asynchronous TDOA Localization With Intermittent Position Aids

We study downlink localization in asynchronous multi-base-station (multi-BS) networks, where time-difference-of-arrival (TDOA) measurements are affected by unknown inter-BS timing offsets and sparse positive excess delays, while absolute position aids are only intermittently available. To address this problem, we formu...

Zi-Xuan Zou, Guo-Chun Luo, Yong-Hua Li · 0 citations
Open access Sep 2026

Adaptive Localization for Underwater Nodes in Uncertain Environments: A Geometric Topology Perception-Enhanced Multi-Stage Reinforcement Learning Strategy

Complex underwater environments induce difficult-to-quantify ranging errors, constraining the localization accuracy and robustness of heterogeneous networks. To address this, a node localization method based on a Geometric Topology Perception-Enhanced Multi-Stage Reinforcement Learning Strategy is proposed. First, an u...

Li-Jun Hao, Chun-Bo Ma, Jian-Bo Cui et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.