MudiNet: A Task-Guided Disentanglement Network for Robust Multipath-Assisted Positioning in Diffuse Environments

Abstract

With the enhancement of signal resolution, multipath components (MPC)s are no longer regarded as noise but rather as valuable information that can contribute to positioning. However, existing research often treats reflective surfaces as ideal reflectors, which is ineffective in handling indistinguishable multipath caused by diffuse reflections. This study approaches diffuse reflectors from the perspective of uncertainty, investigating the statistical properties of indoor diffuse and specular reflections. Based on these insights, a task-guided disentangled representation learning method leveraging multi-epoch channel impulse response (CIR) observations is designed to directly map CIRs to positions, while mitigating the adverse effects of components that contribute minimally to localization accuracy (e.g., diffuse multipath).In this semi-supervised learning framework, a global feature extraction architecture based on self-attention is proposed to capture location-independent wireless environmental information, while an MLP is employed to extract the timevarying features of user equipment (UE) positions. Variational inference based on a latent variable model (LVM) is applied to separate independent features within the CIR, with position labels guiding the LVM to express components more beneficial for localization. Additionally, we provide a feasibility proof for the separability of diffuse and specular environmental features in CIRs. Simulation results demonstrate that the proposed method achieves higher localization accuracy and exhibits stronger robustness against indistinguishable multipath components caused by diffuse scattering. Meanwhile, experimental results on realworld data show that, under a single-input-single-output (SISO) platform, the proposed method attains a localization accuracy of 2.2 m while maintaining comparable model size and inference time. Moreover, compared with simulation settings, the proposed approach yields even larger performance gaps over competing methods in real-world scenarios.