AI Radio Access Network Deployment Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for evaluating the deployment of new radio access equipment in telecommunications networks lack predictive support for decision-making, relying solely on post-deployment observations without providing insights into potential performance impacts before updates.
Innovation Solution
A method utilizing an artificial intelligence module to predict variations in network performance indicators based on historical data and topographical information, enabling the evaluation of candidate configurations and prioritization of deployment strategies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If post-deployment observation methods are used to evaluate network performance, then actual performance data can be obtained, but decision support before deployment is not provided
Solution Approach 1:
The patent applies preliminary action by training an AI prediction model using historical deployment data and performance metrics before actual deployment decisions are made. The model pre-processes topographical information, historical KPIs, and deployment outcomes to create a predictive system that evaluates candidate configurations in advance, enabling informed decisions before resources are committed to actual deployment.
2Ease of operation
If AI prediction models are implemented to forecast performance variations, then decision support before deployment is provided, but system complexity increases
Solution Approach 1:
The patent introduces an intermediary AI prediction model that acts as a mediator between historical data and deployment decisions. This model processes complex topographical features, historical performance data, and deployment configurations to generate simplified performance predictions and rankings, making the decision-making process easier without requiring direct complex analysis by operators.
Solution Approach 2:
The patent creates a virtual copy of the deployment evaluation process through the AI model. Instead of directly analyzing complex real-world deployment scenarios, the model uses historical copies of deployment data and performance metrics to simulate and predict outcomes, reducing the complexity of actual decision-making while maintaining evaluation accuracy.
3Measurement precision
If comprehensive historical data and topographical information are processed, then prediction accuracy is improved, but data processing requirements increase
Solution Approach 1:
The patent extracts only the most relevant features from comprehensive historical data and topographical information for training the AI model. Instead of processing all available data, the system identifies and extracts key predictive features such as historical KPIs, topographical characteristics, and deployment outcomes, reducing data processing requirements while maintaining prediction accuracy.
Data Source
AI summary
A method for evaluating a deployment of a candidate configuration, of radio access equipment of a communication network at a location including: obtaining historical data relating to the network, the historical data comprising topographical information relating to radio access equipment existing in a geographical area before deployment of the candidate configuration at the location level and measurements of one performance indicator of the communication network over an elapsed time period; and predicting a variation of the performance indicator of the communication network, induced by the deployment, the variation being predicted based on the historical network data and topographical information relating to the candidate configuration, and from a previously-learned prediction model, the prediction model being implemented by an artificial intelligence module configured to receive as input data the historical network data and the topographical information of the candidate configuration and to produce as output data the variation of the performance indicator.