Signal Processing Oriented Approach for Big Data Privacy
2015
This paper addresses the challenge of big data security by exploiting signal processing theories. We propose a new big data privacy protocol that scrambles data via artificial noise and secret transform matrices. The utility of the scrambled data is maintained, as demonstrated by a cyber-physical system application. We further outline the proof of the proposed protocol's privacy by considering the limitations of blind source separation and compressive sensing.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
3
References
0
Citations
NaN
KQI