Machine Learning-based xApp for Dynamic Resource Allocation in O-RAN Networks

TL;DR

The paper addresses static resource allocation in 5G RAN by introducing an ML-based xApp within the O-RAN framework that dynamically allocates PRBs and selects the optimal resource allocation policy per base station. It adopts a Random Forest classifier trained on data from a HetNet 5G simulation, deployed within the near-RT RIC, with training managed by the non-RT SMO framework and policy definitions delivered via the A1 interface. Key contributions include a four-policy resource allocation scheme, an ML-driven decision mechanism achieving around 85% policy-classification accuracy, and an evaluation showing improved outage performance and SLA fulfillment. The work demonstrates that ML-driven xApps can enable rapid adaptation of scheduling and resource management in O-RAN, with practical implications for cost-effective, resilient, and high-QoS networks.

Abstract

The disaggregated, distributed and virtualised implementation of radio access networks allows for dynamic resource allocation. These attributes can be realised by virtue of the Open Radio Access Networks (O-RAN) architecture. In this article, we tackle the issue of dynamic resource allocation using a data-driven approach by employing Machine Learning (ML). We present an xApp-based implementation for the proposed ML algorithm. The core aim of this work is to optimise resource allocation and fulfil Service Level Specifications (SLS). This is accomplished by dynamically adjusting the allocation of Physical Resource Blocks (PRBs) based on traffic demand and Quality of Service (QoS) requirements. The proposed ML model effectively selects the best allocation policy for each base station and enhances the performance of scheduler functionality in O-RAN - Distributed Unit (O-DU). We show that an xApp implementing the Random Forest Classifier can yield high (85\%) performance accuracy for optimal policy selection. This can be attained using the O-RAN instance state input parameters over a short training duration.

Figures (6)

• Figure 1: Open Radio Access Network O-RAN Architecture alliance2020ran
• Figure 2: A typical architecture of a 5G network where small and macro cells are deployed to serve multiple user demands
• Figure 3: RIC and xApp Deployment
• Figure 4: Small Cell Network Outage under Different Policies
• Figure 5: ML Training Time