Abstract:Sensing and edge artificial intelligence (AI) are two key features of the sixth-generation (6G) mobile networks. Their natural integration, termed Integrated sensing and edge AI (ISEA), is envisioned to automate wide-ranging Internet-of-Tings (IoT) applications. To achieve a high sensing accuracy, multi-view features are uploaded to an edge server for aggregation and inference using an AI model. The view aggregation is realized efficiently using over-the-air computing (AirComp), which also aggregates channels to suppress channel noise. At its nascent stage, ISEA still lacks a characterization of the fundamental performance gains from view-and-channel aggregation, which motivates this work. Our framework leverages a well-established distribution model of multi-view sensing data where the classic Gaussian-mixture model is modified by adding sub-spaces matrices to represent individual sensor observation perspectives. Based on the model, we study the End-to-End sensing (inference) uncertainty, a popular measure of inference accuracy, of the said ISEA system by a novel approach involving designing a scaling-tight uncertainty surrogate function, global discriminant gain, distribution of receive Signal-to-Noise Ratio (SNR), and channel induced discriminant loss. We prove that the E2E sensing uncertainty diminishes at an exponential rate as the number of views/sensors grows, where the rate is proportional to global discriminant gain. Given channel distortion, we further show that the exponential scaling remains with a reduced decay rate related to the channel induced discriminant loss. Furthermore, we benchmark AirComp against equally fast, traditional analog orthogonal access, which reveals a sensing-accuracy crossing point between the schemes, leading to the proposal of adaptive access-mode switching. Last, the insights from our framework are validated by experiments using real-world dataset.

Explainable AI for Enhancing Efficiency of DL-based Channel Estimation

Towards Explainable AI for Channel Estimation in Wireless Communications

Channel Estimation Based on Deep Learning in Vehicle-to-Everything Environments

ChannelGAN: Deep Learning-Based Channel Modeling and Generating

Deep Learning for Channel Estimation: Interpretation, Performance, and Comparison

Physics-driven AI for Channel Estimation in Cellular Network

Deep Learning Model for Demodulation Reference Signal based Channel Estimation

AI-Based Time-, Frequency-, and Space-Domain Channel Extrapolation for 6G: Opportunities and Challenges.

Explainable Artificial Intelligence (XAI) for 6G: Improving Trust between Human and Machine

Explainable Artificial Intelligence (XAI): An Engineering Perspective

Reconfigurable AI Modules Aided Channel Estimation and MIMO Detection

COST CA20120 INTERACT Framework of Artificial Intelligence Based Channel Modeling

AI enlightens wireless communication: Analyses and solutions for DMRS channel estimation

Deep-Learning-Based Channel Estimation for IRS-Assisted ISAC System

An Explainable Artificial Intelligence Framework for Quality-Aware IoE Service Delivery

Deep Learning Channel Estimation Based on Edge Intelligence for NR-V2I

E-XAI: Evaluating Black-Box Explainable AI Frameworks for Network Intrusion Detection

Deep Learning-based Channel Estimation for Wideband Hybrid MmWave Massive MIMO

Deep Learning for MIMO Channel Estimation: Interpretation, Performance, and Comparison

On the View-and-Channel Aggregation Gain in Integrated Sensing and Edge AI

Explainable AI Using the Wasserstein Distance