Welcome to the AI-Powered 5G Open RAN Optimizer repository!
This repository contains an AI-powered optimization system for Open RAN that leverages machine learning algorithms to optimize the performance of the network. The system can learn from historical data, real-time network data, and external data sources to identify network anomalies, predict network traffic, and optimize network resources allocation.
The AI-powered optimization system includes the following features:
~ Network Anomaly Detection: The system can detect network anomalies and performance degradation in real-time by analyzing network data, such as traffic patterns, signal strength, and congestion.
~ Predictive Network Planning: The system can predict network traffic and plan network resources allocation accordingly to ensure optimal performance.
~ Dynamic Network Optimization: The system can dynamically optimize network resources allocation based on real-time network conditions, such as traffic load, weather conditions, and network congestion.
~ Energy Efficiency Optimization: The system can optimize the energy consumption of the network by dynamically adjusting network resources allocation based on the energy efficiency of different network components.
To implement this project, we used various AI technologies, such as supervised and unsupervised machine learning algorithms, deep learning, reinforcement learning, and natural language processing. We also integrated the AI-powered optimization system with Open RAN interfaces, such as O-RAN, to enable seamless integration and interoperability.
The project is structured as follows:
~ data
: This folder contains the datasets used to train and evaluate the machine learning models.
~ models
: This folder contains the trained machine learning models.
~ notebooks
: This folder contains Jupyter notebooks that were used to develop, test, and analyze the machine learning models.
~ src
: This folder contains the source code for the AI-powered optimization system, including scripts for data preprocessing, training, and inference.
~ tests
: This folder contains unit tests and integration tests for the AI-powered optimization system.
To install the AI-powered optimization system, follow these steps:
- Clone the repository:
git clone https://github.com/N00Bception/AI-Powered-5G-OpenRAN-Optimizer.git
- Install the required dependencies:
pip install -r requirements.txt
To use the AI-powered optimization system, follow these steps:
- Prepare the data:
python src/preprocess.py --input <input_data_path> --output <output_data_path>
- Train the machine learning models:
python src/train.py --input <training_data_path> --output <models_path>
- Evaluate the machine learning models:
python src/evaluate.py --input <evaluation_data_path> --models <models_path>
- Run the AI-powered optimization system:
python src/optimize.py --input <real-time_data_path> --models <models_path> --output <optimized_data_path>
This project was developed by Salim EL GHALBOZURI -Azure AI Engineer- as part of National Institute of Postes and Telecommunications's 5G-RAN Engineer program. We would like to acknowledge the contributions of the Open RAN community and the open-source AI frameworks used in this project.
This project is a work in progress, and we welcome contributions from developers and researchers. If you'd like to contribute, please fork the repository, make your changes, and submit a pull request.
This project is released under the Apache-2.0 License.
This project is still a work in progress. We're actively working on improving the system and adding more features. Please keep in mind that the code is still under development, and we may make breaking changes in the future. We appreciate your understanding and patience.