Scribd
Upload
26 views14 pages

AI For Telecom

The document discusses the potential of AI adoption in telecom, highlighting that 70% of service providers see network planning as the area with the highest return on investment. It emphasizes the use of Reinforcement Learning (RL) for network optimization due to its adaptability in dynamic environments and its ability to operate without explicit modeling. Additionally, it introduces the concept of Intent-Based RAN, which focuses on formal specifications of network expectations to enhance performance during high traffic scenarios, such as events in stadiums.

Uploaded by

yasernajafi76
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
26 views14 pages

AI For Telecom

The document discusses the potential of AI adoption in telecom, highlighting that 70% of service providers see network planning as the area with the highest return on investment. It emphasizes the use of Reinforcement Learning (RL) for network optimization due to its adaptability in dynamic environments and its ability to operate without explicit modeling. Additionally, it introduces the concept of Intent-Based RAN, which focuses on formal specifications of network expectations to enhance performance during high traffic scenarios, such as events in stadiums.

Uploaded by

yasernajafi76
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 14

AI for Telecom

An Operator’s
Perspective
Yaser Najafgholizadeh

NGN & IBS Department

Spring 2025
Overview
Ericsson Survey from Service Providers
• %70 believe the highest potential return from AI adoption will be in their network planning
• %64 percent intend to focus their AI efforts on network performance management.
AI and Learning Types
Which Algorithm is Suitable for Network Optimization?
1. Networks Are Dynamic and Stochastic
• Wireless environments change rapidly:
• users move
• channels vary
• load fluctuates
• RL excels in environments where the rules change
over time and outcomes are uncertain.
• RL learns by trial and error, adapting to these
changes.
2. No Need for Explicit Modeling
• In complex networks (e.g., 5G or heterogeneous
networks), it's hard to model every behavior or
condition explicitly.
• RL doesn't need a perfect model of the environment
—it learns a policy just by interacting. This is
powerful when you're dealing with real-world,
unpredictable conditions.
RL Deployment in Network Optimization
Inside the Agent: A Learning Example

In our example, this model takes as input a measurement of downlink and uplink aggregate radio throughput
and the number of active UE. It then produces a prediction on whether to turn on or off an RF port in a radio
unit when not needed, to save power. Turning off an RF port involves risk, as any traffic on the uplink or
downlink interface that should be received or transmitted from the inactive RF port will also be dropped.
Intent-Based RAN
What are intents?
“the formal specification of all expectations including requirements, goals, and constraints given to a technical system.”

How to use intent based automation to transform networks - Ericsson


Intent-Based RAN
What are intents?
“the formal specification of all expectations
including requirements, goals, and
constraints given to a technical system.”

Scenario: High Traffic During a Stadium Event


Service Intent
• Maintain minimum user throughput of 10 Mbps
• Ensure 95% coverage inside the venue
RAN Intents:
• Ensure throughput ≥ 10 Mbps per user
• Prioritize stadium sector during peak hours
• Allow dynamic spectrum reallocation
Deployed Resources:
• Additional small cells near the stadium
• Temporarily boost existing macro cell power
KPIs:
• Real-time monitoring of throughput, capacity, and coverage
• Deployment Recommendations (from ARMI):
• Suggest deploying 3 small cells and increasing MIMO layers
• Recommend activating mmWave band in the area
Intent-Based RAN
Intent-Based RAN
Multi-layer, intent-based operation – detailed view
Autonomous networks with multi-layer, intent-based operation -
Ericsson
AI based Services in 5G beyond
• Human-machine collaboration

You might also like