Research On Key Technologies Of Location Privacy Protection For Mobile Users

In recent years,with the continuous development of big data,cloud computing,Artificial Intelligence(AI),and communication technology,people have entered a new era of intelligent information.Intelligent mobile devices have been closely connected to people’s lives,and mobile applications are emerging increasingly.Among them,locationbased services(LBS)have been increasingly widely used in people’s daily lives.While location-based services bring such great convenience to people’s lives,they also pose some risks of location privacy leakage.In the process of using location-based services,users need to send their current location information to the service provider.If this location information is abused by the service provider or stolen by a malicious attackers,the user’s personal privacy will be exposed.By collecting and analyzing user’s location information,the adversary can infer personal sensitive information such as home address,company address,income level,interests,hobbies,and movement trajectory.Therefore,protecting the location privacy of mobile users has become a key issue that needs to be addressed for the sustainable and healthy development of location-based services.This dissertation focuses on the application scenarios of location-based services in the Internet of Vehicles(IoV),focusing on the performance and service efficiency of mobile user location privacy protection.This dissertation proposes privacy-preserving method by utilizing system modeling,scheme design,simulation analysis,and other means.The main research work of this dissertation is as follows:1)The problem of location privacy protection is studied in road network scenarios where users are non-uniformly distributed.At the same time,it also considers the performance privacy of query services,and improves service quality by reducing query costs.Joint optimization of privacy protection query costs is modeled as a mixed integer programming problem.Because the problem is non convex,a heuristic feasible solution is given.Minimize query costs and improve the performance of location based services while ensuring privacy and security.According to the weight of users distributed on different road sections,a privacy protection method based on road truncation is proposed.Based on the user density of the road segment,the road segment with a higher user density is truncated and added to the anonymous set.Segments with medium and low user density remain unchanged and are directly added to the anonymous set.In this way,the distribution differences between different road sections are reduced,making it difficult for attackers to distinguish the differences between different road sections,making it more difficult to track users,and achieving privacy protection purposes.Finally,simulation verifies that the design scheme can meet the requirements of k anonymity and l diversity,and can resist edge weight inference attacks in new scenarios.In terms of query services,the scheme proposed in this chapter has lower query costs compared to traditional schemes,and can provide a better Quality of Service(QoS).2)Aiming at the problem of large-scale carpool requests in hot areas,optimization objectives have been formulated to reduce the complexity of the algorithm to ensure that the algorithm can provide matching results in a short time.To improve efficiency,a multi round matching privacy protection method based on road networks is proposed.The map is divided into small areas,filtered based on user itineraries and schedules,and then matched to improve the efficiency of the algorithm by reducing the number of matching users sending alternative sets.At the same time,the scheme also uses ciphertext packaging technology to reduce communication overhead and computational complexity.At the same time,the algorithm is based on a homomorphic encryption scheme,which can calculate the distance between users in the ciphertext space,protecting users’location privacy information from being disclosed.Finally,simulation experiments show that the proposed algorithm consumes less time and can provide accurate matching services within an acceptable time.At the same time,the matching results given in this scheme are more accurate,which can save travel time and reduce overall energy consumption.3)A study was conducted on the location privacy protection of vehicle user broadcast messages in the context of the Internet of Vehicles.Considering that traditional schemes often use static mix zone for user pseudonym changing,which results in poor privacy protection flexibility and the inability of users to send messages during the silent period of pseudonym changing,a group based dynamic mix zone location privacy protection scheme is proposed.This scheme can construct a mix zone at any time and place according to user needs,forming an anonymous set,so that users can change pseudonyms in a timely manner before the expiration of the pseudonym.In addition,during the pseudonym changing,users send emergency security messages to Roadside Unit(RSU)through encryption,and the RSU decrypts the emergency security messages and broadcasts them to the entire network,improving the safety of vehicle driving while ensuring the location privacy of the vehicle user.The simulation results show that the proposed scheme has a lower probability of being tracked compared to the static mix zone scheme,and has a good effect on protecting the location privacy of vehicles,while improving the driving safety of vehicle users.4)The problem of location privacy protection in continuous query services is studied.Using existing trajectory protection schemes for reference,a virtual traj ectory is generated using a generated adversary network to protect the privacy and security of users.In the offline training phase,a confrontation network training center is generated to complete the training.Prepare for model training by collecting real user driving tracks.The collected real trajectory information is preprocessed,classified,and normalized to ensure that the spatiotemporal characteristics of the trajectory are complete,and the driving trajectory can represent a conventional driving traj ectory.After the offline training is completed,the user applies to update the generation module G to generate a virtual trajectory.Security analysis shows that this scheme can protect the user’s location privacy,and attackers cannot obtain an efficient discriminator model to identify the generated virtual reality,effectively protecting trajectory privacy security.Simulation results show that the trajectory generated by the algorithm proposed in this chapter is not easy to identify compared to other schemes,and has better privacy protection performance.At the same time,due to the fact that the training phase is all completed offline,the time consumed to generate the trajectory is extremely short,and it has good usability.