ML-Assisted Admission Control for 5G RAN Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In 5G NR and LTE networks, the random rejection of user equipment (UEs) due to resource unavailability leads to Call Blocking Probability (CBR) and violates Service Level Agreements (SLAs), necessitating an adaptive admission control mechanism to dynamically allocate resources and meet quality of service requirements.
Innovation Solution
A machine learning computing system is integrated into the base station to receive performance indicators, predict traffic parameters for standalone and non-standalone user equipment, and allocate resources accordingly, enabling preemptive actions to adjust resource allocation and operation modes to meet service requirements and reduce call blocking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If random rejection of user equipment is used due to resource unavailability, then resource allocation is simplified, but call blocking probability increases and service level agreements are violated
Solution Approach 1:
The system performs preliminary actions by predicting future traffic parameters using machine learning models before actual resource allocation decisions are made. The ML model forecasts traffic patterns and predicts required resources in advance, enabling the admission control mechanism to proactively allocate resources and prevent call blocking rather than reacting to resource unavailability after it occurs.
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring actual traffic patterns and comparing them with predicted values. The ML model uses historical data and real-time performance indicators to refine its predictions, creating a closed-loop system that adapts to changing network conditions and improves allocation accuracy over time.
2Reliability
If adaptive admission control mechanism is implemented to dynamically allocate resources, then call blocking probability is reduced, but system complexity increases
Solution Approach 1:
The machine learning computing system acts as an intermediary between traffic monitoring and resource allocation functions. Instead of implementing complex adaptive control logic directly in the base station, the ML model serves as a mediator that processes historical data, predicts future traffic patterns, and provides simplified allocation recommendations, reducing the computational complexity burden on the base station while maintaining adaptive control capabilities.
3Measurement precision
If machine learning computing system is integrated into base station to predict traffic parameters, then resource allocation accuracy is improved, but computing resources consumption increases
Solution Approach 1:
The system applies partial action by selectively deploying machine learning predictions only for critical admission control decisions rather than continuously processing all traffic data through complex ML models. The ML model makes predictions at strategic points in the resource allocation process, balancing prediction accuracy needs with computational resource consumption, and avoiding excessive processing for routine operations.
Data Source
AI summary
Systems and methods for radio intelligent controller machine learning assisted admission control are provided. In one example, a method includes receiving performance indicator(s) for standalone UEs and performance indicator(s) for non-standalone UEs from BBU(s) of a base station. The base station includes the BBU(s), a first radio unit, and antenna(s) configured to implement a base station for wirelessly communicating with user equipment in a cell. The method includes determining predicted traffic parameter(s) for standalone UEs based on the received performance indicator(s) for standalone UEs from the BBU(s) and determining predicted traffic parameter(s) for non-standalone UEs based on the received performance indicators for non-standalone UEs from the BBU(s). The method includes allocating resources for standalone and/or non-standalone UEs based on the predicted traffic parameter(s) for standalone UEs, the predicted traffic parameter(s) for non-standalone UEs, and service requirements for the base station.