Numerical information in databases online protecting via random multi-level perturbations

Great research interest is being attracted towards providing unlimited, correct, numerical responses to ad-hoc queries to an on-line database, while not compromising confidential numerical data. Unlike traditional random data perturbation (RDP), a new approach titled multi-level pseudo random sequences covered RDP is carefully extended from our previous research results from pseudo random binary sequences. This paper demonstrates how numerical confidential data can be protected against insider threat. As an application of our method, an online medical survey database is, as a case study, described, which can be easily extended to other application fields, such as e-government and e-commerce. The results from the current paper from multi-level pseudo random sequences compare very well with results obtained from our pseudo random binary sequences.