Journal on Communications ›› 2015, Vol. 36 ›› Issue (Z1): 243-249.doi: 10.11959/j.issn.1000-436x.2015305

• Academic communication • Previous Articles     Next Articles

Privacy protection method on time-series data publication

Dong YU,Hai-yan KANG   

  1. School of Information Management,Beijing Information Science and Technology University,Beijing 100192,China
  • Online:2015-11-25 Published:2015-12-29
  • Supported by:
    The Social Science Foundation of Beijing;The Excellent Talents Program of Beijing;The National Natural Science Foundation of China;Humanity and Social Science Youth Foun-dation of Ministry of Education

Abstract:

A differential privacy model was proposed based on the sampling filtering and the mechanism of evaluation.Firstly,fixed sampling method was used to sample the original data and the non-sampling data be published directly.Secondly,for the sampling date,utilize the differential privacy mechanism to add the noise.Then,use Kalman to correct the sampling date.Finally,use the mutual information to evaluate data under different sampling intervals.Through the experiment,it is proved that the mechanism can achieve a good balance between the practicality and protective.

Key words: differential privacy, time-series data, data publication, sample, Kalman filter, mutual information

No Suggested Reading articles found!