Lightweight Multi-Keyword Ranking Searchable Encryption for Internet of Vehicles

Due to the limited computing and storage capacity of the in-vehicle unit, outsourcing data to a cloud server has become an effective solution to this challenge. However, there is a potential risk of privacy leakage when data is exchanged through cloud servers. Searchable Encryption (SE) solves this problem. Therefore, research on-board units (OBU) querying encrypted data in cloud servers has attracted widespread attention. In our research, we introduced the word2vec model to represent keywords as word vectors. By adopting an improved dual word embedding space model, we successfully reduced the problem of index dimension explosion, allowing lightweight OBU to efficiently handle data query. Through comparative experimental analysis with the traditional multi-keyword ranking searchable encryption (MRSE) scheme, we found that our scheme has achieved significant improvements in both efficiency and accuracy. This research result solves the key problem of low efficiency of vehicle-mounted units due to limited computing and storage capabilities, and provides strong support for the development of ranking searchable encryption in the Internet of Vehicles.