Research On Location Privacy Protection And Measurement Methods For Location-based Services

In recent years,with the rapid development of wireless communication technology and mobile positioning technology,Location Based Service(LBS)based on location technology has been rapidly popularized and widely used,and its application in various industries has gradually deepened.Based on the intelligence of individual consumer needs,location information services will be accompanied by the development of GPS and wireless mobile internet technology,and the demand is showing a significant growth trend.In the field of logistics,the advantages of positioning technology are combined with multiple industrial scenarios,utilizing location data to achieve intelligent management of warehousing and logistics.The application of LBS not only improves the service level of individual or enterprise operations,but also provides more diverse and convenient services for mobile users.Therefore,with the commercialization of LBS services,enormous commercial value is brought,but a large amount of location data is generated.These location data contain relevant privacy information.If location data is obtained by malicious third parties or LBS server terminals during transmission for illegal commercial transactions and profits,the personal user privacy and trade secrets of enterprise users may be leaked,which may cause serious consequences.Once personal or corporate location data is leaked,it can increase their privacy panic and distrust.Enterprises and even the entire LBS service field will experience a significant loss of users,which is not conducive to the healthy development of LBS technology.Therefore,it is necessary to conduct analysis and research on location privacy protection based on location services.In this thesis,the k-anonymity method,dummy location method,encryption method,and the location privacy measurement method in the location privacy protection are studied.Moreover,the above methods is applied and analyzed in the scenario of intelligent logistics system vehicle delivery in practice.The specific research work is as follows:1.K-anonymity is the most widely used method for location privacy protection,but the spatial region is affected by the geographical environment,anonymous user distribution,etc.,and it is difficult to achieve a better privacy protection effect only by various regular geometric regional anonymity methods.In view of the shortcomings of spatial anonymous region construction methods,a polygon k-anonymous location privacy protection method based on density distribution is proposed.The idea of k-anonymity is applied in this method to construct an irregular convex polygon anonymous region,which can better match the terrain environment.At the same time,according to the density distribution of k-value,the ideal and effective anonymous region is determined.The location privacy protection method combining spatial anonymity and dummy location method is used to implement location privacy protection.The strength of privacy protection and the quality of query service are effectively balanced,and the effect of k-anonymity location privacy protection is better realized.2.Aiming at the issue of location privacy leakage caused by semantic attacks in location privacy protection methods based on dummy location,an approximate matching dummy location selection method is proposed.Firstly a region is divided into m×m grids in this method,and all locations are divided into different grids.Then,according to the distribution of all locations in different grids,the Morton code of each location point is calculated.By approximate matching calculation,the dummy location candidate set is generated by selecting dummy locations located on non adjacent grids with the minimum distance between these grids.Then,for the location points in the candidate set,the semantic similarity between any two points is calculated through their placename information,and the k-1 location points with the largest semantic similarity are selected as dummy location.The k locations including the real location and k-1 dummy locations are sent to the LBS server for query.In this method,the preprocessing process is simplified,the physical dispersion and semantic diversity of dummy locations are ensured,and the efficiency of dummy location generation is improved.3.Encryption is the best method to protect the location privacy,but the query efficiency is relatively low,and the sorting of query results in the ciphertext state is not realized in most existing methods.In order to solve the problems of location data generation time is long,query efficiency is low,and query results cannot be sorted in the ciphertext state in encryption based location privacy protection methods,a location privacy protection method based on ciphertext retrieval is proposed.In this method,the spatial location coordinates are converted into binary strings,and the encryption scheme combining public key encryption and scrambling encryption is adopted.The location data is encrypted and then outsourced to the cloud server.On the server side,the nearest neighbor location query and top-k sorting of query results in ciphertext state are implemented.Experimental results show that the location privacy protection effect is guaranteed in this method,the generation time of location data is shortened,and the efficiency of ciphertext query is effectively improved.4.Aiming at the lack of universally applicable performance evaluation methods in privacy protection performance analysis,a location privacy measurement model based on entropy weight method is proposed.Firstly,the algorithm performance evaluation indicators are extracted from the security framework and application system of location privacy protection.Considering the characteristics of anonymity,location semantics,and physical dispersion of location,the weights are configured according to the privacy level and location privacy protection indicators.Then,the entropy weight method is used to evaluate and measure the effect of location privacy protection,and combined with the idea of game theory,the optimized comprehensive entropy weight value is obtained.Experimental results show that the effectiveness of location privacy protection can be more effectively evaluated in this method.By analyzing and evaluating the weight of location privacy indicators,an optimization model for location privacy protection based on geographical semantics is proposed,and the effectiveness of this method is verified through experiments.5.In the intelligent logistics service system for intelligent mobile terminal applications,the location privacy protection method proposed in this thesis is applied and tested in the privacy protection scenario of logistics delivery vehicles.In this scheme,the intelligent logistics system is taken as the scene to protect the location privacy of vehicle terminals.The location privacy protection technology based on polygon k-Anonymous,semantic similarity,and ciphertext retrieval is deployed and applied on the Android platform.This technology aims to protect the location privacy of intelligent terminal users,and at the same time,location-based applications can be used normally.Through application in actual location service scenarios,the application effects of the several methods proposed in this thesis are tested,and its performance advantages are analyzed.Furthermore,the feasibility and effectiveness of this scheme are verified.