The Design And Implementation Of A Random Response System Based On Homomorphism

With the development of economy and society, people’s material life has been improving rapidly, spirituality and self-awareness have also promoted. Therefore, the exclusion to social survey involving privacy has increased too. However, the result of some investigations on the social management offers a great reference value for the public benefits or academic study. So the social investigation involving privacy is necessary. The survey mentioned above is a sensitive issue, which involves personal privacy, such as drug addiction, extramarital affair or some other unknown personal habits. When the survey is referred to privacy issues, the respondent may be give an answer which does not meet the actual situation with the consideration of self-protection. In a word, it is difficult to obtain the real result for such investigation with traditional methods. The reason is that we can’t protect the respondents’ privacy. So this kind of survey is a difficult spot in social survey.This paper presents a novel protocol to solve the problem mentioned above, which is a Randomized Response Protocol Based on Homomorphism encryption (we will call the system RRPBH for short). This protocol can effectively compute the sum of respondents’answers simultaneously to protect their privacy. The main content and framework of this paper are as follows:(1) Targeted random surveyWe conduct a survey which is pointed to the two selected sensitive questions, namely the survey questions only have two answers:yes or no. In our survey, we use1and0to represent yes and no respectively.(2) Application of the Paillier algorithmThe Paillier algorithm has the good properties of addition homomorphism and mixed multiplicative homomorphism. Homomorphic encryption is a kind of cryptography based on the computational complexity of mathematical problem. After homomorphic encryption, we can get an output, then this output will be decrypted, the result after decrypted is equal to the result used the same method to compute the original data. We used this good property in our protocol. This homogeneity on the addition is the most important pillar in our protocol.(3) Effective protection for data transmissionIn our RRPBH system, investigators need to transmit lots of information to respondents. At the same time, respondents also need to share a lot of information according to a certain order. In the process of information transmission, in order to prevent respondents’malicious tamper with information, we require the investigator sign his digital name on the information, and the information should include the public key of his digital signature, so that each respondent can verify the message.