SPOT: Single-Shot Positioning via Trainable Near-Field Rainbow Beamforming

2025-11-14

AI Summary

Achieving high-accuracy, low-overhead user localization in broadband sensing and positioning under a single downlink transmission remains challenging. Method: This paper proposes an end-to-end deep learning framework that jointly optimizes reconfigurable intelligent surface (RIS) phase shifts and true-time-delay coefficients to synthesize task-oriented, frequency-dependent near-field rainbow beams; it further introduces a lightweight neural network that directly decodes 2D angle–distance coordinates from quantized received power feedback. Contribution/Results: The key innovation lies in treating both phase and time-domain beamforming parameters as trainable variables—enabling joint beam design and localization while breaking the single-transmission bottleneck. Compared with state-of-the-art approaches, the method reduces system overhead by an order of magnitude and significantly improves both accuracy and robustness in 2D localization error. It establishes a new paradigm for low-overhead, high-precision broadband positioning.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Phase-time arrays, which integrate phase shifters (PSs) and true-time delays (TTDs), have emerged as a cost-effective architecture for generating frequency-dependent rainbow beams in wideband sensing and localization. This paper proposes an end-to-end deep learning-based scheme that simultaneously designs the rainbow beams and estimates user positions. Treating the PS and TTD coefficients as trainable variables allows the network to synthesize task-oriented beams that maximize localization accuracy. A lightweight fully connected module then recovers the user's angle-range coordinates from its feedback of the maximum quantized received power and its corresponding subcarrier index after a single downlink transmission. Compared with existing analytical and learning-based schemes, the proposed method reduces overhead by an order of magnitude and delivers consistently lower two-dimensional positioning error.

Problem

• Simultaneously designs rainbow beams and estimates user positions
• Maximizes localization accuracy through trainable beamforming coefficients
• Reduces positioning overhead and error with single downlink transmission

Innovation

• Phase-time arrays integrate phase shifters and true-time delays
• End-to-end deep learning designs rainbow beams and estimates positions
• Lightweight module recovers coordinates from single downlink transmission