Unsupervised Learning for AoD Estimation in MISO Downlink LoS Transmissions

2025-03-15

AI Summary

Accurate Angle-of-Departure (AoD) estimation in MISO downlink line-of-sight (LoS) channels faces fundamental trade-offs among estimation accuracy, latency, and privacy, as existing methods rely heavily on large-scale labeled datasets and real-time channel observations. Method: This paper proposes, for the first time in the SLAC (Simultaneous Localization and Communication) context, an unsupervised end-to-end deep learning framework that unifies deterministic (DML) and stochastic (SML) estimation paradigms within a single, input-agnostic neural architecture. The method jointly exploits received signal and pilot sequence features without requiring ground-truth labels or empirical measurements. Contribution/Results: Experiments demonstrate that the approach reduces required observations by over 60%, while significantly lowering communication overhead and latency. It consistently outperforms state-of-the-art supervised and conventional algorithms in AoD estimation accuracy, establishing a lightweight, high-accuracy, low-latency, and privacy-preserving localization paradigm for IoT devices.

Technology Category

• Machine Learning: Hardware-aware ML
• Intelligent Robots: State Estimation
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Security and Privacy: Large-scale security measurements

Abstract

With the emerging of simultaneous localization and communication (SLAC), it becomes more and more attractive to perform angle of departure (AoD) estimation at the receiving Internet of Thing (IoT) user end for improved positioning accuracy, flexibility and enhanced user privacy. To address challenges like large number of real-time measurements required for latency-critical applications and enormous data collection for training deep learning models in conventional AoD estimation methods, we propose in this letter an unsupervised learning framework, which unifies training for both deterministic maximum likelihood (DML) and stochastic maximum likelihood (SML) based AoD estimation in multiple-input single-output (MISO) downlink (DL) wireless transmissions. Specifically, under the line-of-sight (LoS) assumption, we incorporate both the received signals and pilot-sequence information, as per its availability at the DL user, into the input of the deep learning model, and adopt a common neural network architecture compatible with input data in both DML and SML cases. Extensive numerical results validate that the proposed unsupervised learning based AoD estimation not only improves estimation accuracy, but also significantly reduces required number of observations, thereby reducing both estimation overhead and latency compared to various benchmarks.

Problem

• Estimates Angle of Departure (AoD) for MISO downlink LoS transmissions
• Reduces real-time measurement needs for latency-critical applications
• Minimizes data collection for training deep learning models

Innovation

• Unsupervised learning for AoD estimation
• Unified training for DML and SML methods
• Reduced observations and estimation latency