Information Density as a Quantitative Measure for AI-enabled Virtual Sensing: Feasibility and Limits

2026-05-05

AI Summary

This study addresses the limitations of conventional IoT sensing and compression methods, which suffer from low computational efficiency and irreversible information loss, thereby hindering scalable and efficient virtual sensing. To overcome these challenges, the authors propose an AI-driven virtual sensing framework that introduces a novel two-dimensional information density metric based on phase feature space and mutual information. By jointly leveraging spatiotemporal and cross-modal correlations, the framework enables high-fidelity perception without physical sensors. It further supports optimal sensor placement selection within and across modalities. Validated on real-world data from the Madrid smart city deployment, the approach achieves virtual sensing with an average error below 3.21% using only a single physical sensor, substantially enhancing system efficiency and scalability.

Abstract

Modern IoT and sensor networks generate vast amounts of data, posing significant challenges for storage, transmission, and real-time processing. Traditional approaches, such as compressive sensing and machine learning-based compression, often suffer from computational inefficiencies and irreversible data loss. This paper introduces Information Density as a quantitative metric to support sensor deployment and enable AI-driven virtual sensing. We propose a framework that leverages spatial, temporal and inter-modal correlations among sensor signals to perform sensing tasks even in the absence of physical sensors. Two complementary measures: (i) Phase in Eigen Space and (ii) Mutual Information, are developed to quantify and assess information density, enabling the selection of optimal sensor configurations across both intra-modality and cross-modality scenarios. Validated using real-world data from Madrid's smart city infrastructure, this framework demonstrates the feasibility of replacing physical sensors with virtual ones under bounded error conditions (e.g., achieving $<3.21\%$ mean error with a single sensor). The results highlight the potential for scalable and energy-efficient sensing systems in smart environments.

Problem

• Information Density
• Virtual Sensing
• Sensor Networks
• IoT
• Data Efficiency

Innovation

• Information Density
• Virtual Sensing
• Mutual Information
• Phase in Eigen Space
• AI-enabled Sensing