Exponential Moving Maximum Filter for Network Peak Data Smoothing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network reporting tools fail to adequately perform trend prediction, especially with peak resource usage data that includes significant fluctuations, as traditional predictive analytics techniques do not handle irregular data well.
Innovation Solution
The implementation of an Exponential Moving Maximum (EMM) filter, which smoothes data series while maintaining local maximum values, is applied to network-related data to facilitate more accurate trend prediction by preserving peak values and overcoming the time frame boundary effect.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional predictive analytics techniques are used on peak resource usage data, then the analysis process is simple, but the prediction accuracy deteriorates due to significant fluctuations and irregular data patterns
Solution Approach 1:
The EMM filter is applied as a preliminary processing step before predictive analytics. By pre-smoothing the peak resource usage data while preserving local maximum values, the filter prepares the data in advance to be more suitable for prediction algorithms, thereby improving prediction accuracy without adding complexity during the main analysis phase
Solution Approach 2:
The EMM filter acts as an intermediary between the raw fluctuating data and the predictive analytics system. It transforms the irregular peak resource usage data into a smoothed sequence that maintains important peak characteristics, serving as a bridge that enables accurate predictions without requiring the prediction algorithm to directly handle noisy raw data
2Stability of the object's composition
If data smoothing is applied to reduce fluctuations, then data irregularity is reduced, but local maximum values may be lost or distorted
Solution Approach 1:
The EMM filter applies different treatment to different parts of the data sequence. For each data point, it calculates the exponential moving maximum based on previous values, giving special attention to preserving local maximum values while smoothing other portions. This localized approach ensures that peak values maintain their accuracy while the overall data becomes smoother
Solution Approach 2:
The filter uses an exponential weighting parameter (alpha) that controls the balance between smoothing and preserving peak values. By adjusting this parameter, the system can dynamically change the degree of smoothing applied while maintaining the integrity of local maximums, thus achieving both data smoothness and peak value accuracy
Data Source
AI summary
Techniques for an exponential moving maximum (EMM) filter for predictive analytics in network reporting are disclosed. In some embodiments, a process for predictive analytics in network reporting using an EMM filter includes pre-processing network-related data by performing exponential moving maximum (EMM) filtering on the network-related data; and determining predictive analytics based on the EMM filtered network-related data.