Out-of-Distribution Detection in Wireless Multimodal Foundation Models for 6G ISAC

2026-07-23

AI Summary

In integrated sensing and communication for 6G, wireless multimodal foundation models are vulnerable to out-of-distribution (OOD) data, leading to silent failures that compromise system trustworthiness. This work proposes WMFM-OOD, a novel framework that uniquely combines metric-based geometric prototyping with temperature-scaled probabilistic scoring. Specifically, it constructs base station prototypes in a joint latent space to capture the intrinsic manifold structure of valid wireless environments and employs a temperature scaling mechanism to calibrate output probabilities, thereby effectively distinguishing in-distribution samples from those affected by covariate shift. Evaluated on the DeepVerse6G dataset, the method achieves an AUROC of 0.8824 and reduces the false positive rate at 95% true positive rate (FPR95) by approximately 17% compared to uncalibrated baselines, significantly enhancing both robustness and system usability.

Abstract

The integration of Foundation Models (FMs), such as the Wireless Multimodal Foundation Model (WMFM), into 6G networks provides a unified framework for Integrated Sensing and Communication (ISAC), leveraging generalized representations to simultaneously optimize data transmission and environmental perception. However, the deployment of such data-driven models in safety-critical infrastructure is hindered by the Out-of-Distribution (OOD) problem, which poses a fundamental threat to system trustworthiness. Standard FMs operate under a closed-world assumption, rendering them vulnerable to silent failures when deployed in unseen radio environments. To address this reliability gap and ensure trustworthy network operation, we propose WMFM-OOD, a robust metric-based OOD detection framework. Unlike traditional methods that rely on raw compatibility scores, WMFM-OOD constructs geometric Base Station (BS) Prototypes within the joint latent space to capture the manifold structure of valid radio environments. By employing a temperature-scaled probabilistic scoring mechanism, our approach effectively distinguishes between In-Distribution (ID) and covariate-shifted anomalies. We validate the framework on the DeepVerse6G dataset. Experimental results demonstrate that WMFM-OOD significantly outperforms uncalibrated baselines, achieving an Area Under the Receiver Operating Characteristic Curve (AUROC) of 0.8824 and reducing the False Positive Rate (FPR) at 95 % True Positive Rate (TPR), commonly referred to as FPR95, by approximately 17% in the optimal temperature regime, thereby providing an initial layer of detection sensitivity to mitigate catastrophic model failures without completely disrupting network availability.

Problem

• Out-of-Distribution Detection
• Wireless Multimodal Foundation Model
• 6G ISAC
• Model Trustworthiness
• Covariate Shift

Innovation

• Out-of-Distribution Detection
• Wireless Multimodal Foundation Model
• Integrated Sensing and Communication
• Geometric Prototypes
• Temperature-scaled Scoring