Along with an integration of wearable devices, wireless communications and big data in the smart health, biomedical data is collected referring to multiple associated patients during interactions. Due to communication channel openness and data sensibility, privacy preservation become increasingly noteworthy in the edge and cloud hybrid computing based healthcare applications. In this work, a cooperative privacy preservation scheme is designed for wearable devices with identity authentication and data access control considerations in the space-aware and time-aware contexts. In the space-aware edge computing mode, secret sharing and MinHash based authentication is designed to enhance privacy preservation along with similarity computing without revealing sensitive data; In the time-aware cloud computing mode, ciphertext policy attribute based encryption is applied for fine-grained access control, and bloom filter is used to achieve efficient data structure without privacy exposure. The GNY logic based security formal analysis is performed to prove theoretical correctness, and the proposed scheme achieves cooperative privacy preservation for wearable devices in smart health with communication overhead and computation cost.

In order to meet low latency, service sensitive and location awareness requirements of digital manufacturing, fog computing is introduced to be an intermediate layer between industrial Internet of Things (IoT) and cloud. The distributed, dynamic characteristics and the collaboration requirement make it face many new security and privacy issues that cannot be solved by the traditional public key or symmetric cryptosystem. For addressing them, a registered but anonymous attribute credential is designed to manage the network entities. Based on it, an attribute credential based public key cryptography (AC-PKC) is constructed to provide flexible key management by taking the advantage of the certificate-less public key cryptography (CL-PKC) and the combination property of the elliptic curve cryptography(ECC). Encryption, authentication and access control with privacy preserving can be realized on the basic operations of AC-PKC, which can meet various security requirements of fog computing based digital manufacturing. The performance analyses and comparison with the existing public key schemes and attribute based encryption(ABE) solutions show that the proposed scheme can work flexibly at a relative low cost.

Point cloud is an important 3D data structure, but its irregular format brings great challenges to deep learning. The advent of PointNet makes it possible to process irregular point cloud data by neural networks directly. As an extension of PointNet, PointNet++ can extract local features, which makes it perform better than PointNet in processing point cloud data. But in practice, it is common that the density of a point set usually varies with the location, which makes the computation overhead of PointNet++ very heavy. To deal with it, we propose an octree grouping-based network structure for PointNet++, named Octree-Grouping-PointNet++ (OG-PointNet++). It determines the point density by constructing an unbalanced octree for the point cloud, and groups point according to the point density. These point groups are assigned to different layers according to their density, and the local feature of each group is extracted by PointNet++. The global feature is obtained from the last abstract layer and used for classification and segmentation. The experiments show its competitive performance in many 3D tasks, such as object classification and semantic segmentation.