Reliable near-field source counting and positioning with controlled risk
Reliable Near-Field Source Enumeration and Positioning Informed by Two-Stage MUSIC
Information TheoryMachine Learning
Summary
It can be hard to find and locate multiple signal sources nearby when reflections confuse the receiver. The authors designed a method that carefully counts how many sources exist without missing any, and uses a deep learning approach inspired by a classic signal method to pinpoint source locations directly. They also provide a way to measure how sure we are about each location estimate. Their tests show better accuracy and trustworthy predictions than earlier methods.
What this means in practice
- •For wireless system engineers: Improve near-field multi-source positioning accuracy and reliability in wireless communication setups experiencing signal reflections and interference.
- •For indoor navigation developers: Use enhanced source counting and positioning to build more reliable indoor localization systems that cope with multipath signals.
Authors
Jiaying Li, Haifeng Wen, Changsheng You, Yuanwei Liu, Hong Xing
Abstract
Accurate and reliable near-field multi-source localization is essential for location-aware processing in future wireless systems. However, coherent multipath propagation complicates the direct recovery of line-of-sight (LoS) source positions, while existing estimation methods generally lack explicit control of the risk of source underestimation and positioning uncertainty. In this paper, we investigate reliable source enumeration and positioning in near-field coherent multipath environments. Specifically, we develop a conformal risk control (CRC) based source enumeration scheme that calibrates the penalty coefficient of an eigenvalue-based criterion to control the expected miss-detection ratio at a prescribed level. We further propose MUSIC-Net, a model-based deep learning framework informed by two-stage MUltiple SIgnal Classification (MUSIC), which learns the LoS-related subspace and surrogate distances through subspace-distance-based training, thereby enabling direct source positioning without estimating non-line-of-sight (NLoS) path parameters or performing subsequent path association. To quantify the uncertainty of the estimated positions, we develop a split conformal prediction (SCP) based calibration method that constructs angle-range predicted regions with finite-sample simultaneous coverage guarantees. Numerical results show that the proposed framework effectively controls the risk of source underestimation, while providing improved positioning accuracy with tighter predicted regions than benchmarks given the same target coverage levels.