Terminology » Near-Real-Time RIC vs Non-Real Time RIC: O-RAN Intelligence Demystified

Near-Real-Time RIC vs Non-Real Time RIC: O-RAN Intelligence Demystified

5g near real time ric non real time ric ric ran o-ran

Introduction: Within the O-RAN architecture, the RAN Intelligent Controller (RIC) is split into two major domains viz. Non-Real Time RIC and Near-Real Time RIC. The non RT RIC handles longer term analytics, policies and model training, while the near RT RIC responds quickly to changing radio conditions with fine grained control. They are very essential for implementing AI/ML driven radio optimization.

Non-RT RIC

Non-Real Time RIC (Non-RT RIC) operates in the management and orchestration layer of the O-RAN architecture, typically within the Service Management and Orchestration (SMO) framework. It functions on a time scale greater than one second and focuses on long-term network optimization, policy management, analytics, and AI/ML model training.

The Non-RT RIC provides guidance and policies to the Near-RT RIC through the A1 interface, helping shape how near-real time control decisions are made. Its main role is to enable smarter, data driven network behavior by learning from historical trends, user patterns, and performance data to improve efficiency, capacity, and user experience across the network.

Near-RT RIC

Near-Real Time RIC (Near-RT RIC) operates closer to the radio layer and handles network optimization and control in near real time (10 ms to 1 second). It interacts directly with O-DUs and O-CUs via the E2 interface, running applications called xApps that perform dynamic radio resource management, mobility control, interference mitigation and load balancing.

The Near-RT RIC allows the network to adapt quickly to changing conditions, such as user movement or traffic spikes, improving spectral efficiency and quality of service. Together, it bridges the gap between AI-driven policy decisions from the Non-RT RIC and time critical radio control at the edge.

Key Differences

Aspect	Non-Real-Time RIC (Non-RT RIC)	Near-Real-Time RIC (Near-RT RIC)
Timescale/Latency	Operates on slower control loops (>=1 sec)	Faster Control Loops (Approx. 10 ms to 1 seconds)
Main function	Long term optimization, analytics, model training (AL/ML) , rApps that generate strategies etc.	Real time/near real time execution of policies from non-RT RIC, xApps that do load balancing, resource allocation, interference mitigation etc.
Deployment location	Typically centralized, part of SMO, often in central cloud or central data centers	Deployed closed to edge or regional cloud/ near RAN/ near DU/CU nodes in order to meet tight latency requirements
Interfaces used	A1 , O1, O2, internal interfaces for rApps	E2, A1, internal messaging/xApps framework
Scope	Broad network or multi-cell scope	More localized
Compute requirements	Often more compute needed for heavy data processing, ML model training, large scale analytics	Needs to be efficient for quick decision making
Responsibility	Focus: defines, refines policies, long term strategies	Focus: executes or enacts decisions that follow policies

Conclusion: The non-RT RIC sets strategic direction with large scale analytics and policies and the near-RT RIC enacts those policies with low latency actions in the RAN. Together, they form the intelligence backbone of O-RAN, enabling automation, adaptation and efficient network operations in 5G and beyond.