Data-driven Method to Ensure Cascade Stability of Traffic Load Balancing in O-RAN Based Networks
Mengbang Zou, Yun Tang, Weisi Guo
TL;DR
This work addresses cascade stability in O-RAN load balancing by learning data-driven load dynamics from real RU data and deriving a stability condition that guarantees convergence to a balanced state. It employs a library-based dynamic identification (sparse regression) to capture self and offloading dynamics, yielding $ rac{d l_i}{dt}=oldsymbol{ heta}(l_i)oldsymbol{\xi_i}$, and uses a Gershgorin-based analysis to confirm stability around the equilibrium $l_1=\cdots=l_N=1$ via a matrix $oldsymbol{K}=\mathbf{F}'(1)+\mathbf{Q}-\mathbf{A}\odot\mathbf{P}$. The RIC is then equipped to monitor RU dynamics and adjust load distribution in real time to prevent cascade handovers and Ping-Pong effects. Simulation results on a 12-RU O-RAN setup demonstrate that the approach identifies accurate per-RU self-dynamics, achieves convergence to the desired load distribution, and outperforms non-stability-aware policies in terms of cascade stability and QoS. The practical impact lies in more reliable, energy-efficient RAN operation with scalable stability guarantees across arbitrary network topologies and heterogeneous RU dynamics.
Abstract
Load balancing in open radio access networks (O-RAN) is critical for ensuring efficient resource utilization, and the user's experience by evenly distributing network traffic load. Current research mainly focuses on designing load-balancing algorithms to allocate resources while overlooking the cascade stability of load balancing, which is critical to prevent endless handover. The main challenge to analyse the cascade stability lies in the difficulty of establishing an accurate mathematical model to describe the process of load balancing due to its nonlinearity and high-dimensionality. In our previous theoretical work, a simplified general dynamic function was used to analyze the stability. However, it is elusive whether this function is close to the reality of the load balance process. To solve this problem, 1) a data-driven method is proposed to identify the dynamic model of the load balancing process according to the real-time traffic load data collected from the radio units (RUs); 2) the stability condition of load balancing process is established for the identified dynamics model. Based on the identified dynamics model and the stability condition, the RAN Intelligent Controller (RIC) can control RUs to achieve a desired load-balancing state while ensuring cascade stability.