Network Utilization Estimation Using Segmented Time Series Regression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Service providers face challenges in accurately estimating network utilization, leading to overcharging or underutilization of network resources, as existing methods like linear regression models fail to account for changes in trends and seasonality, resulting in inaccurate billing and resource allocation.
Innovation Solution
A composite regression model is employed, which includes a linear regression model and a second model that accounts for seasonality, allowing the identification of segments in time series data, each associated with a separate regression model, to determine an estimated network utilization based on the current segment's model, thereby improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a simple linear regression model is used to estimate network utilization, then the device complexity is reduced, but the measurement precision deteriorates due to inability to account for trends and seasonality
Solution Approach 1:
The patent segments the time series data into multiple segments based on identified change points, where each segment is modeled by a separate linear regression model. This allows the system to capture local trends and seasonality patterns without requiring a single complex global model, thus resolving the contradiction between model complexity and measurement precision.
Solution Approach 2:
The patent implements a dynamic modeling approach where the regression model parameters can change at identified change points in the time series data. This dynamic adaptation allows the system to respond to changing network utilization patterns while maintaining relatively simple local linear models, balancing complexity and accuracy.
2Measurement precision
If a composite regression model with multiple segments is used to account for trends and seasonality, then the measurement precision improves, but the device complexity increases
Solution Approach 1:
The patent divides the time series data into multiple segments separated by change points, with each segment having its own simple linear regression model. This segmentation strategy improves measurement precision by capturing local patterns while avoiding the complexity of a single comprehensive model, as each segment can be modeled independently with straightforward linear regression.
Solution Approach 2:
The patent applies partial action by using simple linear regression models for each segment rather than attempting to model all patterns globally. This approach achieves sufficient precision for each local segment without the excessive complexity of a unified complex model, resolving the contradiction between precision improvement and complexity increase.
3Adaptability or versatility
If change points are identified to segment time series data, then the adaptability improves for capturing local patterns, but the difficulty of detecting and measuring increases
Solution Approach 1:
The patent uses feedback mechanisms where the identified change points inform the selection of subsequent regression models for each segment. This feedback loop allows the system to adapt to local patterns effectively, as each segment's model is informed by the change point detection from previous segments, improving adaptability while managing detection difficulty through iterative refinement.
Data Source
AI summary
In one implementation, a method includes obtaining time series data. The time serious data includes a plurality of network utilization measurements. The plurality of network utilization measurements is indicative of a plurality of utilizations of one or more resources of a network resource at a plurality of times. The method also includes determining whether the time series data comprises a plurality of segments. Each segment of the plurality of segments is associated with a separate regression model and each segment includes a portion of the time series data. The method further includes identifying a current segment from the time series data when the time series data comprises the plurality of segments. The method further includes determining an estimated network utilization based on a current regression model associated with the current segment.