Mobile network agents share control when goals are compatible
Understanding Conflict and Compatibility Conditions for Agents in AI-RAN
Networking and Internet Architecture
Summary
Mobile networks often have multiple agents trying to control the same parts of the system, which can cause conflicts if their goals clash. The authors propose a new way to check if these agents can work together without causing problems, based on understanding their target goals and acceptable ranges. They tested their method in simulated networks and found it helps agents share control smoothly when their objectives align, while still handling conflicts well when they don’t. This can improve how mobile networks are managed automatically.
What this means in practice
- •For mobile network operators: Enable multiple control agents to jointly manage network parameters only when their performance goals can be met simultaneously, reducing unnecessary conflicts.
- •For telecommunications equipment developers: Build network management systems with smarter conflict resolution that only intervenes when agent control objectives truly conflict, improving automation efficiency.
Authors
Arshia Zolghadr, Joao F. Santos, Imtiaz Nasim, Deniz Aytemiz, Nicholas J. Kaminski, Jacek Kibilda
Abstract
Disaggregated mobile networks expose open interfaces for independent agents to control the Radio Access Network (RAN). The interactions between agents with contrasting objectives can result in conflicts. However, existing conflict mitigation methods treat agents that control the same parameter as being in a conflict that requires arbitration of their actions, without assessing whether their objectives can be satisfied simultaneously. In this paper, we introduce a control-theoretic approach to determine when agents can coexist and share control of the same parameters. We model threshold-based agents often encountered in the literature as sampled deadband feedback controllers and represent their objectives through Key Performance Indicator (KPI) targets and tolerances that define acceptable deadband bounds. We derive compatibility conditions from these bounds and show that compatibility for any number of agents and monitored KPIs can be established through pairwise tests. We propose a conflict mitigation policy that allows compatible agents to share control of the same network parameters while restricting arbitration to cases where their objectives cannot be satisfied simultaneously. We evaluate our policy through simulations using NIST's ns3-oran extension of ns-3, and our results show that it enables agents to share control with near-zero Service Level Agreement (SLA) drift when their objectives are compatible and the shared acceptable range is attainable, while matching Priority Mitigation when their objectives are incompatible.