Privacy-preserving Smart Parking Navigation Supporting Efficient Driving Guidance Retrieval

Abstract

It is frustrating and time consuming for drivers to find an available parking spot in a congested area, such as downtown and shopping malls, especially in peak hours. Thus, it is very helpful for drivers to have real-time parking information to assist them in finding vacant parking spots timely. Unfortunately, to acquire needed parking information, the drivers have to submit personal queries for the availability of parking spaces in their destinations, and this could result in privacy violation if the queries are not protected. To reduce drivers’ hassle and preserve drivers’ privacy, we propose a privacy-preserving smart parking navigation system (P-SPAN) with efficient navigation result retrieval for drivers using Bloom filters. P-SPAN enables a cloud to guide vehicles to vacant parking spaces in the destinations based on real-time parking information without disclosing any personal information about drivers. Specifically, an efficient data retrieval mechanism is developed based on Bloom filters to support navigation result retrieval for querying vehicles. The drivers can anonymously query accessible parking spots to the cloud, and efficiently retrieve the encrypted navigation results from the passing-by roadside units. Therefore, it is unnecessary for a vehicle to keep connected with the queried roadside unit for acquiring the navigation result. Performance evaluation demonstrates that P-SPAN can provide effective parking navigation with high navigation result retrieving probability and low computational and communication overhead.