Home / Current Issue / Paper 1723131
A Cognitive Capacity-Orchestration Model for Energy Efficient Optical Transport Networks Under Dynamic Telecommunications Traffic
Subject area: Science,Engineering and Technology · Area of research: Dynamic Telecommunications Traffic
Abstract
Telecommunications traffic is no longer predictable. With 5G, video streaming, and cloud services, traffic fluctuates within minutes. Traditional optical transport networks run at fixed capacity to handle “peak load”, which means they waste a lot of energy during low traffic period operators are simultaneously under pressure to cut power consumption because electricity is one of their biggest opex costs. This paper proposes a “cognitive capacity-orchestration Model”. This means that it is an AI-driven system that thinks about current and predicted traffic, then dynamically turns up or down the capacity of optical links, transponders, and amplifiers to save energy without breaking service quality. The cognitive part means it learns the traffic patterns and makes decisions automatically. The goal is to balance two things: energy efficiency and service continuity under dynamic traffic. The scale of optical networks demands an advanced automation frameworks capable of adapting to dynamic service requirements, multi-vendor environments, and physical-layer impairments. Large Language Models (LLMs), which is built on transformer architectures, offer a paradigm shift by enabling context-aware reasoning, natural language interpretation, and multi-task generalization. These models can automate configuration generation, alarm correlation, fault diagnosis, routing and spectrum assignment (RSA), while enhancing Quality of Transmission (QoT) estimation and scenario modeling. Optical networks serve as the foundation of modern communication systems, supporting high-bandwidth, long reach data transmission through light signals propagated in optical fiber media. Optical networks support a wide range of services, including internet backbones, cloud structure and 5G connectivity. A typical optical network comprises of components such as transponders (which convert electrical signals to optical ones), reconfigurable optical add-drop multiplexers, which directs optical signals, erbium-doped fiber amplifiers, and optical fibers that carry the signal across the physical layer.
How to cite this paper
@article{1723131,
author = {Samuel Okoro},
title = {A Cognitive Capacity-Orchestration Model for Energy Efficient Optical Transport Networks Under Dynamic Telecommunications Traffic},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {10},
number = {3},
pages = {4037-4042},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1723131.pdf},
abstract = {Telecommunications traffic is no longer predictable. With 5G, video streaming, and cloud services, traffic fluctuates within minutes. Traditional optical transport networks run at fixed capacity to handle “peak load”, which means they waste a lot of energy during low traffic period operators are simultaneously under pressure to cut power consumption because electricity is one of their biggest opex costs. This paper proposes a “cognitive capacity-orchestration Model”. This means that it is an AI-driven system that thinks about current and predicted traffic, then dynamically turns up or down the capacity of optical links, transponders, and amplifiers to save energy without breaking service quality. The cognitive part means it learns the traffic patterns and makes decisions automatically. The goal is to balance two things: energy efficiency and service continuity under dynamic traffic. The scale of optical networks demands an advanced automation frameworks capable of adapting to dynamic service requirements, multi-vendor environments, and physical-layer impairments. Large Language Models (LLMs), which is built on transformer architectures, offer a paradigm shift by enabling context-aware reasoning, natural language interpretation, and multi-task generalization. These models can automate configuration generation, alarm correlation, fault diagnosis, routing and spectrum assignment (RSA), while enhancing Quality of Transmission (QoT) estimation and scenario modeling. Optical networks serve as the foundation of modern communication systems, supporting high-bandwidth, long reach data transmission through light signals propagated in optical fiber media. Optical networks support a wide range of services, including internet backbones, cloud structure and 5G connectivity. A typical optical network comprises of components such as transponders (which convert electrical signals to optical ones), reconfigurable optical add-drop multiplexers, which directs optical signals, erbium-doped fiber amplifiers, and optical fibers that carry the signal across the physical layer.},
month = {September},
}