A Multi-Stakeholder Perspective on Self-Managing Networks
Patrick Weber, Artur Sterz, Bernd Freisleben, Oliver Hinz
TL;DR
The paper argues that self-managing networks should adopt a multi-stakeholder perspective to handle increasing complexity and information asymmetries beyond the traditional operator-centric view. It presents a system model for identifying stakeholders, their preferences, and unified utility functions, while addressing conflicts of interest and information asymmetries with signaling, monitoring, and game-theoretic approaches. Through three case studies—D2D data transfer, energy-efficient IoT multi-hop, and edge computing placement—it demonstrates substantial performance gains and cost reductions achievable under a multi-stakeholder framework. The work highlights practical implications for network design and policy, and outlines future directions for fostering collaboration, transparency, and real-world validation of multi-stakeholder self-managing networks.
Abstract
Modern telecommunication networks face an increasing complexity due to the rapidly growing number of networked devices and rising amounts of data. The literature advocates for self-managing networks as a means to tackle the resulting challenges. While self-managing networks provide potential solutions to these challenges, current research solely focuses on the perspective of network operators. However, modern telecommunication networks involve various stakeholders, such as service providers and end users, and necessitate interactions between them. By transitioning from a single-stakeholder to a multi-stakeholder perspective, we address the preferences of all involved parties, acknowledging potential conflicts of interest and constraints like information asymmetries. This broader perspective facilitates the development of more effective self-managing networks, significantly enhancing their performance metrics compared to approaches that solely prioritize the concerns of network operators.