Mixture of Experts-augmented Deep Unfolding for Activity Detection in IRS-aided Systems

2025-02-27

AI Summary

Device activity detection (DAD) in intelligent reflecting surface (IRS)-assisted massive machine-type communication (mMTC) under hybrid channel fading remains challenging; existing methods rely on single-fading assumptions and require prior knowledge of channel types. Method: We propose a model-driven, mixture-of-experts (MoE)-enhanced deep unfolding network. Its gating mechanism dynamically selects specialized expert subnetworks tailored to distinct channel characteristics, while integrating projected gradient descent with an accurate IRS channel model to achieve channel-agnostic robust detection. Contribution/Results: Experiments demonstrate that the proposed method reduces misdetection rate by 32% over covariance-based baselines and black-box deep neural networks under hybrid fading. It significantly improves generalization and detection accuracy, and—crucially—enables the first adaptive DAD without requiring prior channel-type information.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Intelligent Robots: State Estimation
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

In the realm of activity detection for massive machine-type communications, intelligent reflecting surfaces (IRS) have shown significant potential in enhancing coverage for devices lacking direct connections to the base station (BS). However, traditional activity detection methods are typically designed for a single type of channel model, which does not reflect the complexities of real-world scenarios, particularly in systems incorporating IRS. To address this challenge, this paper introduces a novel approach that combines model-driven deep unfolding with a mixture of experts (MoE) framework. By automatically selecting one of three expert designs and applying it to the unfolded projected gradient method, our approach eliminates the need for prior knowledge of channel types between devices and the BS. Simulation results demonstrate that the proposed MoE-augmented deep unfolding method surpasses the traditional covariance-based method and black-box neural network design, delivering superior detection performance under mixed channel fading conditions.

Problem

• Enhances activity detection in IRS-aided systems
• Combines deep unfolding with mixture of experts
• Improves performance under mixed channel fading

Innovation

• Mixture of Experts framework
• Deep unfolding technique
• Automatic channel type selection