Resampling Multi-Resolution Signals Using the Bag of Functions Framework: Addressing Variable Sampling Rates in Time...

Area of Science:

• Time Series Analysis
• Signal Processing
• Data Science

Background:

• Accurate time series analysis requires handling data with varying sampling rates.
• Traditional methods often necessitate uniform sampling frequencies, limiting their applicability.
• Irregularly sampled data is common in finance, healthcare, and IoT networks.

Purpose of the Study:

• To introduce the Multi-Resolution Bag of Functions (MR-BoF) framework for time series analysis.
• To develop a method that accommodates signals with differing resolutions and sampling rates.
• To demonstrate the framework's effectiveness in data reconstruction and resampling.

Main Methods:

• The MR-BoF framework utilizes sampling-rate-independent techniques for time series decomposition.
• A flexible encoding approach integrates multi-resolution time series data.
• Experiments were conducted to validate the framework's performance.

Main Results:

• The MR-BoF framework enables precise reconstruction of original time series data.
• The method enhances resampling capabilities by leveraging decomposed signal components.
• Significant advantages were observed in scenarios with irregular sampling rates.

Conclusions:

• The MR-BoF framework offers a robust solution for analyzing time series data with heterogeneous sampling rates.
• This approach is valuable for applications in finance, healthcare, industrial monitoring, and sensor networks.
• The framework provides a flexible and accurate tool for modern data analysis challenges.