Decentralized AI Service Placement, Selection and Routing in Mobile Networks

2025-11-04

AI Summary

To address the traffic pressure and user mobility challenges induced by large-scale AI models in mobile networks, this paper proposes a decentralized AI service co-optimization architecture. We formulate a joint optimization problem—integrating service deployment, selection, and request routing—that is tightly coupled and mobility-constrained; to solve it, we develop a nonlinear queuing-delay model and apply KKT condition analysis to derive an analytical optimization framework. We further design a distributed Frank–Wolfe algorithm with a lightweight message protocol, enabling efficient non-convex optimization while avoiding costly service migrations. Additionally, a traffic tunneling mechanism ensures service continuity during user mobility. Experimental results demonstrate that our approach achieves a superior trade-off between end-to-end latency and quality of service, significantly outperforming conventional MEC solutions—particularly by relaxing their restrictive assumptions on network topology and mobility patterns.

Technology Category

• Search and Optimization: Distributed Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Optimization

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

The rapid development and usage of large-scale AI models by mobile users will dominate the traffic load in future communication networks. The advent of AI technology also facilitates a decentralized AI ecosystem where small organizations or even individuals can host AI services. In such scenarios, AI service (models) placement, selection, and request routing decisions are tightly coupled, posing a challenging yet fundamental trade-off between service quality and service latency, especially when considering user mobility. Existing solutions for related problems in mobile edge computing (MEC) and data-intensive networks fall short due to restrictive assumptions about network structure or user mobility. To bridge this gap, we propose a decentralized framework that jointly optimizes AI service placement, selection, and request routing. In the proposed framework, we use traffic tunneling to support user mobility without costly AI service migrations. To account for nonlinear queuing delays, we formulate a nonconvex problem to optimize the trade-off between service quality and end-to-end latency. We derive the node-level KKT conditions and develop a decentralized Frank--Wolfe algorithm with a novel messaging protocol. Numerical evaluations validate the proposed approach and show substantial performance improvements over existing methods.

Problem

• Optimizing AI service placement and routing in mobile networks
• Balancing service quality and latency under user mobility
• Addressing limitations of existing mobile edge computing solutions

Innovation

• Decentralized framework jointly optimizes placement, selection, routing
• Traffic tunneling supports mobility without service migrations
• Decentralized Frank-Wolfe algorithm with messaging protocol