Aiming at the efficiency and privacy of queries under mobile edge computing, secure skyline query processing in mobile edge computing over location-based data is developed. A secure skyline query framework in the mobile edge computing scenario is first proposed and a security model is also defined. To improve the efficiency of secure queries in the edge server, we use the Paillier homomorphic encryption and order-revealing encryption to devise a novel and unified secure index structure. It can effectively protect the query unlinkability in accordance with the semi-blind structure proposed by us. Furthermore, based on the secure index, we propose a secure skyline query protocol in mobile edge computing to realize the edge-cloud collaborative skyline query considering the privacy leakage. Finally, we analyze the complexity and security of the proposed approach. Findings from the experimental evaluation show that our proposed approach significantly increases the efficiency under the semi-honest model, in comparison to the state-of-art.