Fixed Wireless Network Performance Prediction via Dynamic Model Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high cost and complexity of deploying and maintaining large dense fixed wireless networks, particularly due to factors like loss of line of sight and weather conditions, necessitate a predictive solution for network performance degradation, while existing prediction methods lack dynamic adaptability and user-friendly interfaces for real-time monitoring.
Innovation Solution
A method and system that dynamically selects the best prediction model for network performance based on user preferences for approximation speed and accuracy, using a light-weight user interface for real-time monitoring and backend processing, incorporating machine learning for data analysis and model selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple prediction functions are used to improve prediction accuracy, then the prediction precision is improved, but the device complexity increases
Solution Approach 1:
The system dynamically selects prediction functions based on runtime conditions and user preferences rather than using a fixed set of functions. This allows the system to adapt its complexity level according to current needs, choosing simpler functions when speed is prioritized and more complex functions when accuracy is prioritized, thus resolving the contradiction between prediction accuracy and system complexity
Solution Approach 2:
The system changes parameters such as the selection of prediction functions, time ranges, and data sources based on user preferences for speed and accuracy. By adjusting these parameters dynamically, the system can optimize the balance between prediction accuracy and computational complexity for different operational scenarios
2Measurement precision
If complex prediction models are used to improve prediction accuracy, then the prediction precision is improved, but the processing time increases
Solution Approach 1:
The system dynamically adjusts the complexity of prediction models based on user preferences and runtime conditions. When users prioritize speed, the system selects simpler prediction functions that execute faster; when users prioritize accuracy, the system selects more complex functions. This dynamic adaptation resolves the trade-off between processing time and prediction accuracy
Solution Approach 2:
The system applies partial action by selecting only the necessary prediction functions and time ranges needed for the current prediction task based on user preferences. Instead of always running all available complex models, the system executes only the appropriate subset, reducing processing time while maintaining required accuracy levels
3Reliability
If real-time network monitoring is implemented to improve network reliability, then the reliability is improved, but the use of energy increases
Solution Approach 1:
The system implements partial monitoring by focusing on specific prediction targets and time ranges rather than continuous full-network monitoring. Users can specify particular time ranges and performance metrics to monitor, allowing the system to consume less energy while still providing reliable predictions for critical parameters
Solution Approach 2:
The system changes monitoring parameters such as prediction time ranges, data sampling frequencies, and alert thresholds based on user preferences and network conditions. This allows the system to maintain network reliability for critical parameters while reducing energy consumption by adjusting or suspending monitoring of less critical parameters during low-risk periods
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods and systems to predict network performance of a fixed wireless network is disclosed. In one embodiment, a method is performed by an electronic device, the method comprises receiving data of a fixed wireless network; modeling the data using a plurality of functions, where the modeling comprises identifying a prediction function that provides a best approximation of the data based on a user preference of approximation speed and accuracy; predicting a network performance value in a future time using the prediction function; and notifying when the predicted network performance value fails to meet a threshold value.