We propose a privacy-preserving proximity based location query (PPLQ) protocol, which enables each querier to get the location information of publisher without violating users location privacy.The proposed protocol utilizes prefix membership verification and predicate encryption technique to achieve the secure multi-dimensional keyword search and private bilateral proximity testing simultaneously.We propose a novel technique, multi-scale technique to present users location coordinates. This method enables each user to flexibly set his own location policy.We evaluate the efficacy and efficiency of the proposed scheme by using real-world dataset. The security analysis and performance evaluation show that the proposed protocol is secure and efficient. Location based services are increasingly popular, partly due to the trend of smartphone and online social network service adoption. However, it is important for location-based service provider (LBSP) to ensure user location privacy in the provision of such services. In this paper, we present a secure hitch service in location based social networks (LBSNs). To provide such a service, we propose a privacy-preserving proximity based location query (PPLQ) protocol, which is based on the hierarchical predicate encryption technique and the prefix membership verification technique. There are two types of users in this system, namely: the querier and the publisher. Our protocol allows a querier to query the location of publishers using multi-dimensional search, and it enforces distance based access control in the location queries. In order to improve the efficiency of our protocol, we use the multi-scale technique to represent users location information in the query condition and searchable index. The proposed protocol is designed to achieve multi-dimensional keyword search and bilateral private proximity testing simultaneously. Our protocol enables each user to independently define his/her own location policy for private proximity testing. In particular, we propose some solutions to reduce the search time cost of the CSP so that the time cost is acceptable for queriers. Finally, we demonstrate the utility of the protocol using simulated data on the map of the city area of Changsha and a U.S. census dataset.