An association probability based noise generation strategy for privacy protection in cloud computing


Zhang, Gaofeng; Zhang, Xuyun; Yang, Yun; Liu, Chang; Chen, Jinjun

Available versions


Cloud computing allows customers to utilise IT services in a pay-asyou- go fashion to save huge cost on IT infrastructure. In open cloud, 'malicious' service providers could record service data from a cloud customer and collectively deduce the customer's privacy without the customer's permission. Accordingly, customers need to take certain actions to protect their privacy automatically at client sides, such as noise obfuscation. For instance, it can generate and inject noise service requests into real ones so that service providers are hard to distinguish which ones are real. Existing noise obfuscations focus on concealing occurrence probabilities of service requests. But in reality, association probabilities of service requests can also reveal customer privacy. So, we present a novel association probability based noise generation strategy by concealing these association probabilities. The simulation comparison demonstrates that this strategy can improve the effectiveness of privacy protection significantly from the perspective of association probability.

Publication year


Publication type

Conference paper


Lecture notes in computer science: 10th International Conference on Service-Oriented Computing (ICSOC 2012), Shanghai, China, 12-15 November 2012 / Chengfei Liu, Heiko Ludwig, Farouk Toumani and Qi Yu (eds.), Vol. 7636, pp. 639-647






Copyright © 2012 Springer-Verlag Berlin Heidelberg. The accepted manuscript is reproduced in accordance with the copyright policy of the publisher. The definitive version of the publication is available at