Internet of things (IoT) has been increasingly applied to kinds of new network environments. In the ciphertext access control of IoT, it is required to achieve fine-grained access control policy on the authorized entities. Unfortunately, the traditional communication channel security cannot well satisfy the security and privacy requirements from the “one-to-many” and “many-to-many” scenarios such as resisting the secret key sharing attack. Privacy preserving outsourced computation in IoT can realize secure data aggregation and signal processing in the encrypted domain, protect the individual data privacy and the aggregation result privacy, and verify the correctness of computation. However, most of the existing work exploited (fully) homomorphic encryption directly on data themselves, incurring inefficiency and significantly deviating from the principle of hybrid encryption. To address the issues outlined above, in this paper, we introduce recent concrete solutions exploiting cryptographic techniques, especially presenting the following two results proposed by our research team: a new mechanism of traceable and revocable attribute-based encryption and a new method of lightweight privacy preserving aggregating n data by executing the public key encryption only once. On the meanwhile, we also give a survey on the privacy preserving techniques in popular network services such as smart grid, wireless body area network and wireless vehicular ad hoc network, and suggest the recent advances and results with respect to the challenging open problems in this research field all over the world.