Quantifying information content in data compression using the autocorrelation function

S. Sanborn*, X. Ma

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

This letter presents an algorithm that uses time-series statistical techniques to analyze the information content changes of an individual signal time series subjected to data compression. Autocorrelation-based methods are used to analyze the residuals between the original and processed signals. T-statistic analysis of Autocorrelation Function (ACF) coefficients is used to determine an Upper Specification Limit (USL) of the data compression ratio. The approach provides a continuous data quality measurement that is useful in datalogging systems design and online performance monitoring of such systems. The proposed algorithms can be widely applied to remote monitoring and diagnosis systems. A practical example is also described in the letter.

Original languageEnglish
Pages (from-to)230-233
Number of pages4
JournalIEEE Signal Processing Letters
Volume12
Issue number3
DOIs
Publication statusPublished - 1 Mar 2005

Keywords

  • Autocorrelation function
  • Compression efficiency
  • Data compression
  • Data quality metric
  • Information content analysis
  • Monitoring and diagnostics

Fingerprint Dive into the research topics of 'Quantifying information content in data compression using the autocorrelation function'. Together they form a unique fingerprint.

Cite this