Home > Archive > 2013 > Volume 3 Number 5 (Oct. 2013) >
IJMLC 2013 Vol.3(5): 408-412 ISSN: 2010-3700
DOI: 10.7763/IJMLC.2013.V3.350

Multi-Scale 1-Bit Compressed Sensing Algorithm and Its Application in Coalmine Gas Monitoring System

Xu Yonggang, Zhang Yi, and Hua Gang

Abstract—It is imperative to reduce load of the underground cable channel in coming Mining of Things for thousands of sensors. Gas data, for example, acts an essential role in mass time-continuous data sets of coalmine monitoring systems, we propose a multi-scale 1-bit compressive sensing algorithm in this paper to effectively compress data according the statistical properties ‘regular pattern’ of gas data sequence. The algorithm divides input into non-uniform intervals according to the prior attention of the gas monitoring information, then signal decision threshold and the compressed scales depend on the different attention in order to achieve a large scales of compression ratio on redundant data and as much as possible to maintain the sensitive information, comparing with the traditional 1-bit compressive sensing which brings overload quantization distortion during uniform quantization. Satisfactory results obtained by simulation and actual field applications show, which provides a useful reference to similar real monitoring data compression acquisition with compressed sampling.

Index Terms—Mining of things (MoT), multi-scale 1-bit compressed sensing (MS 1-bitCS), coalmine safety.

Xu Yonggang and Hua Gang are with School of Information & Electronic Engineering, China University of Mining and Technology, Xuzhou, Jiangsu, 221008, China (e-mail: feilongxyg@163.com, ghua3323@163.com). Zhang Yi is with Sience & Thechnology Center, Huaibei Coalmine Grp. Ltd., Huaibei, Anhui, China (e-mail: zy@hbcoal.com).


Cite:Xu Yonggang, Zhang Yi, and Hua Gang, "Multi-Scale 1-Bit Compressed Sensing Algorithm and Its Application in Coalmine Gas Monitoring System," International Journal of Machine Learning and Computing vol.3, no. 5, pp. 408-412, 2013.

General Information

  • ISSN: 2010-3700 (Online)
  • Abbreviated Title: Int. J. Mach. Learn. Comput.
  • Frequency: Bimonthly
  • DOI: 10.18178/IJMLC
  • Editor-in-Chief: Dr. Lin Huang
  • Executive Editor:  Ms. Cherry L. Chen
  • Abstracing/Indexing: Inspec (IET), Google Scholar, Crossref, ProQuest, Electronic Journals Library.
  • E-mail: ijmlc@ejournal.net

Article Metrics