Year: 2026
Type: preprint
Abstract: Industrial processes generate vast amounts of timeseries data, necessitating efficient compression methods to reduce storage and transmission costs while preserving data accuracy. This project introduces a time-series compression algorithm tailored to industrial data, leveraging pattern recognition to identify and compress repetitive sequences. The proposed method not only enhances compression efficiency but also ensures the retention of critical information required for analysis and forecasting. Experimental results demonstrate that our approach achieves a 5 to 6.5 times higher compression ratio compared to traditional compression algorithms, with minimal loss of data fidelity. This makes it a scalable and effective solution for real-time industrial data management.
Source: Research Square
Institution: PES University
Language: English
Subfield: Signal Processing
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7236727645
