Research On Location Privacy Protection In Ride-hailing Service

In recent years,with the rapid development of the shared economy,online ride-hailing platforms such as Uber,Lyft and Didi have sprung up.They combine the traditional taxi industry with the Internet to provide convenient,fast,safe and comfortable travel services for millions of users by taking advantage of modern information technology.This form of service allows users to book a taxi in advance before departure,reducing his waiting time.Meanwhile,the driver can also receive the order farther away from the passenger and then drive to the passenger's place of departure,reducing the driver's probability of unloaded driving and improving the driver's travel efficiency.However,in order to provide travel services accurately,on time and quickly,online ride-hailing platform need to collect a large amount of sensitive information about drivers and passengers,including the driver's status(whether in idle mode or busy mode),location,servicing time,and the passenger's location and travel time.If these information are obtained by an unauthorized adversary,passengers or drivers may be subject to location-based deception,which in extreme cases could threaten the safety of passengers or drivers'property or life.Therefore,it is necessary for us to protect the privacy of passengers' location when providing online ride-hailing services.Aiming at the privacy leakage problems existing in the existing online ride-hailing system,the current research on location privacy protection algorithms mainly includes space cloaking-based,K-anonymity-based,pseudonym-based,differential privacy-based,homomorphic encryption-based and blockchain-based,but these algorithms have some shortcomings.The distance between the passenger and the driver in the location privacy protection algorithms based on space cloaking is a Euclidean distance between the center of the two cloaking space,not the actual traveled distance,the driver may need to travel a longer distance than the actual distance to pick up passengers,so that the efficiency of the online ride-hailing system is not achieved.The location privacy protection algorithms based on K-anonymity are not able to withstand the attack of background knowledge.The query processing cost of the location privacy protection algorithms based on pseudonymous is relatively high.The location privacy protection algorithms based on differential privacy is difficult to allocate privacy budget.The location privacy protection algorithms based on the homomorphic encryption not only discloses the driver's location privacy,but also increases the calculation overhead of the passenger.Most of the location privacy protection algorithms based on blockchain only discusses the feasibility of using blockchain to provide privacy protection for online ride-hailing services,but does not propose scene implementation details of blockchain-based online ride-hailing service system.Aiming at the above challenges,this paper analyzes the problem of location privacy leakage in online ride-hailing services,and puts forward three solutions which make up for some shortcomings of the existing schemes.Specific research contents are as follows:1.An efficient location privacy protection method based on local sensitive hash algorithm is proposed.This scheme replaces the GPS coordinates of the user's exact location with the interest points around the passenger,and converts the distance between the passenger and the driver into the coincidence of the two sets of interest points.And then the distance between the passenger and the driver can be calculated quickly by taking advantage of Minhash algorithm.This method greatly improves the computational efficiency,and effectively protects the user's location privacy;2.Location privacy protection method based on public key encryption scheme with equality test is proposed.Considering that the public key encryption scheme with equality test allows authorized individuals or institutions to judge whether two different ciphertexts correspondings to the same plaintext information without decryption.Therefore,under the condition that the online ride-hailing service platform is not fully trusted,this paper designs a matching scheme of online ride-hailing service with privacy protection.In this method,passengers and drivers can encrypt their location information,and the ride-hailing platform can match drivers and passengers in the form of ciphertext.3.A matching method between passengers and drivers based on blockchain and public key encryption scheme with equality test is proposed.This paper puts forward an online ride-hailing service system based on blockchain technology,eliminates the intermediaries in the current online ride-hailing service system,and uses the blockchain technology and smart contracts to build a decentralized online ride-hailing service platform,overcoming the fact that traditional taxis are organized and operated by third parties and cannot resist the single point of failure to attack.