Research On The Protection Method Of Sensitive Information In Meteorological Dat

Meteorological data is a valuable resource.A large number of meteorological data provide support for meteorological services and meteorological research,however,these meteorological data often contain a lot of sensitive information that affects national security,social stability and personal privacy.Designing a reasonable scheme to protect the security of sensitive information of meteorological data not only plays an important role in safeguarding national security,maintaining social stability and preventing the leakage of personal sensitive information,but also provides guarantee for meteorological research and meteorological information sharing.This paper introduces the sensitive information contained in meteorological information.A new random k-anonymity method is designed,which overcomes the disadvantage that the traditional k-anonymity can not resist exhaustive attack;A new differential privacy method is designed,which reduces the amount of added noise and improves the availability of data sets;Based on the above two methods,schemes to protect the location privacy and trajectory privacy in meteorological crowdsourcing activities are designed.Firstly,a new random K-anonymous implementation method based on adding noise and randomization is proposed.This method can improve the strength of privacy protection,reduce the amount of information loss and improve the availability of data set.The random k-anonymity method proposed in this paper makes records indistinguishable by randomization and adding noise,and thus,K-anonymity is realized.This method does not require records that in the same anonymous group to have the same quasi identifier.The attacker cannot judge that those records belong to the same anonymous group,so it can resist exhaustive attack.Compared with the traditional K-anonymous,this method has higher execution efficiency,less damage to data and less information loss.Secondly,a differential privacy method is proposed,which reduces the amount of added noise and improves the availability of protected data.Differential privacy can protect privacy effectively,however,adding too much noise will cause great damage to the original data.This paper proposes a new differential privacy implementation method,which adds less noise to achieve differential privacy,thus reducing the damage of adding noise to the original data set and improving the availability of the data set.Thirdly,based on polar transformation and differential privacy,a location privacy protection method suitable for crowdsourcing is proposed.This method can reduce the travel distance of crowdsourcing workers and improve the success rate of crowdsourcing task allocation.Crowdsourcing is a popular working mode at present,which has been widely used in meteorological research and achieved good results.This paper proposes a location privacy protection method that can be used in meteorological crowdsourcing.This method can not only protect the location privacy of crowdsourcing workers and crowdsourcing tasks,but also ensure a high success rate of task allocation.Fourthly,through time slicing and random sampling,a trajectory privacy protection method based on differential privacy technology is proposed.This method greatly improves the efficiency of privacy protection method without reducing the intensity of privacy protection.In the process of meteorological investigation and research,the trajectory of meteorological researchers and meteorological crowdsourcing workers may disclose relevant sensitive information.This paper proposes a trajectory privacy protection method,which is based on differential privacy and random sampling.This method does not depend on the structure of trajectory data,and can be applied to the privacy protection of arbitrary trajectory data.Through time slicing and random sampling,the efficiency of differential privacy transformation is greatly improved and the intensity of privacy protection is enhanced.