Research On Wechat User Geolocation Technology Based On Reported Information Feature

Location Based Social Networks(LBSNs)represented by WeChat are profoundly changing people's way of life and social interaction,but at the same time,it provides convenience for criminals to carry out illegal activities.WeChat user geolocating technology take advantage of the target user's reported information provided by WeChat to determine the location of the target user.WeChat user geolocating technology has wide application demand and development prospect in sensitive target geolocating,network public opinion monitoring,administrative law enforcement and other fields.However,influenced by privacy protection strategies in WeChat,there is a complex correspondence relationship between the actual distance and the reported distance,which leads to the difficulty of geolocating target users reliably in the actual environment.In this thesis,by statistical analysis of the relationship between the reported distance and the actual distance in the WeChat "People Nearby" service and the relative order among users in the query result list,three kinds of WeChat user geolocating algorithms are proposed.The main work of this thesis is as follows:1.In view of the existing WeChat user geolocation algorithms are easy to misjudge in the process of space partition,a WeChat user geolocation algorithm based on the statistical characteristics of the reported distance is proposed.Firstly,test environment is built to collect a large number of reported data.Based on the statistical characteristics of the actual distance distribution corresponding to each reported distance,the target reported distance and the corresponding range of actual distance are selected.Secondly,the potential area is scanned by probes to find the target user and the initial space where the target user is located is determined.Thirdly,probes are deployed based on the stepwise strategy,and the subspace of the target user is determined by analyzing the change of the target users' reported distance collected by different probes.Finally,the deployment strategy of the probe is adjusted based on the current space range,and the location of the target user is realized by partition the current space continually.The results of actual geolocating experiment for WeChat users show that,compared with the classical spatial partition based geolocation algorithm and the heuristic number theory based geolocation algorithm,the average geolocating accuracy of the proposed algorithm is higher.2.For the problem that the existing WeChat user geolocation algorithm is difficult to get the actual distance range of target user based on the reported distance,a geolocating algorithm based on the missequence in query result list is proposed.The algorithm takes advantage of the attribute of WeChat “People Nearby” service that there exists relative order between users in the query result list.Firstly,the hybrid interval that lead to missequence is determined by statistical analysis of the relationship between the users' relative order and the actual distance to the probe.Secondly,the distance between the reference probe and the probe is dynamically adjusted to make the relative order between the reference probe and the target user in probe's query result is changed.The hybrid interval is used to determine the actual distance range between the target user and the probe.Finally,the location of the target user is achieved based on the trilateration method.The geolocating experiment for 50 WeChat users is carried out.The experimental results show that,compared with the existing heuristic number theory based geolocation algorithm and advanced trilateration based geolocation algorithm,the proposed algorithm can achieve higher geolocation accuracy.3.To address the problem that the existing WeChat user geolocation algorithm is high-overhead in the process of geolocating,a geolocation algorithm based on the sequence change of users in the query result list is proposed.Firstly,the reference probe is constructed and the binary search method is used to adjust the distance between the reference probe and the probe,so that the reference probe is adjacent to the target user in the query result of the probe.Secondly,the change array of the actual distance between reference probe and the probe is obtained by judging the adjacent relationship between the target user and the reference probe in the query result.Based on the array,the actual distance between the target user and the probe is determined.Finally,the location of the target user is geolocated based on the trilateration method.According to the geolocating experiments for WeChat users,the proposed algorithm can achieve higher geolocating accuracy with lower overhead compared with the existing heuristic number theory based geolocation algorithm.Finally,the whole work of the thesis is summarized and the problems which are worthy for further research are pointed out.