Spatial Signal Design for Positioning via End-to-End Learning

TL;DR

This paper introduces an end-to-end learning approach using an autoencoder architecture to optimize transmitter and receiver design for mmWave positioning, achieving comparable or superior results to traditional methods especially under hardware impairments.

Contribution

It presents a novel autoencoder-based framework for joint optimization of signal design and positioning in mmWave systems, demonstrating its effectiveness over model-based methods.

Findings

End-to-end learning matches model-based design in ideal conditions.

The proposed method outperforms traditional approaches under hardware impairments.

Autoencoder architecture effectively estimates UE position with multiple BSs.

Abstract

This letter considers the problem of end-to-end learning for joint optimization of transmitter precoding and receiver processing for mmWave downlink positioning. Considering a multiple-input single-output (MISO) scenario, we propose a novel autoencoder (AE) architecture to estimate user-equipment(UE) position with multiple base-stations (BSs) and demonstrate that end-to-end learning can match model-based design, both for angle of departure (AoD) and position estimation, under ideal conditions without model deficits and outperform it in the presence of hardware impairments.