Cellular Network Traffic Prediction via External Event Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current telecommunication networks face challenges in predicting increases in traffic, particularly due to human crowds and events, which can lead to network performance degradation and require significant time and resources to manage.
Innovation Solution
A method and node in a telecommunication network that receive event information with geographic and temporal data from external sources, associate this information with access nodes, and predict traffic increases to adjust parameters such as emission power and antenna tilt, enabling efficient network optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If network parameters are adjusted manually in response to traffic increases, then network performance can be optimized, but significant time and resources are required for management
Solution Approach 1:
The system receives event information from external sources and predicts future traffic increases before they occur. By associating events with access nodes based on geographic data and coverage areas, the system proactively identifies cells that will experience traffic increases, allowing parameter adjustments to be made in advance rather than reactively after traffic spikes occur.
Solution Approach 2:
The system enables automatic self-optimization by having access nodes automatically adjust their own parameters (emission power, antenna tilt) based on predicted traffic increases from associated events. This automation eliminates the need for manual intervention and reduces the time and resources required for network management while maintaining optimal performance.
2Productivity
If network parameters are adjusted proactively based on event prediction, then network operation effectiveness is improved during events, but system complexity increases
Solution Approach 1:
The system introduces an intermediary prediction layer that receives event information from external sources, processes it through association rules and geographic matching, and generates traffic increase predictions. This intermediary layer acts as a bridge between external event data and internal network parameter control, automating the decision-making process and reducing the complexity of direct manual management.
Solution Approach 2:
The system creates a multi-functional prediction mechanism that handles multiple event types, geographic associations, and parameter adjustments through a unified approach. By developing a general-purpose event-to-traffic prediction system that can be applied across different access nodes and event scenarios, the solution achieves high productivity without proportionally increasing complexity.
3Measurement precision
If event information is collected from external sources, then traffic increase prediction accuracy is improved, but information acquisition complexity increases
Solution Approach 1:
The system employs an intermediary information processing layer that receives diverse event information from multiple external sources, standardizes the data formats, and processes it through association rules. This intermediary layer simplifies the complexity of collecting and processing external information by creating a unified interface between external event data and internal prediction algorithms, thereby improving prediction accuracy without proportionally increasing acquisition complexity.
Data Source
AI summary
A method for predicting an increase in amount of traffic in a particular cell of a telecommunication network, wherein said telecommunication network comprises a plurality of access nodes, wherein each of said access nodes is arranged to serve a cell in said telecommunication network, wherein each cell covers a predefined coverage area, said method comprising the steps of receiving event information from at least one source which is external to said telecommunication network, wherein said event information comprises geographic data and temporal data of events that are to take place, associating said events with at least one particular access node of said plurality of access nodes based on said geographic data of each of said events and based on said predefined coverage areas of said cells and predicting an increase in amount of traffic in a cell of said telecommunication network based on said associated events with at least one particular access node and their corresponding temporal data.