Due to the strong analytical ability of big data, deep learning has been widely applied to model on the collected data in industrial Internet of Things (IoT). However, for privacy issues, traditional data-gathering centralized learning is not applicable to industrial scenarios sensitive to training sets, such as face recognition and medical systems. Recently, federated learning has received widespread attention, since it trains a model by only sharing gradients without accessing training sets. But existing research works reveal that the shared gradient still retains the sensitive information of the training set. Even worse, a malicious aggregation server may return forged aggregated gradients. In this article, we propose the VFL, a verifiable federated learning with privacy-preserving for big data in industrial IoT. Specifically, we use Lagrange interpolation to elaborately set interpolation points for verifying the correctness of the aggregated gradients. Compared with existing schemes, the verification overhead of VFL remains constant regardless of the number of participants. Moreover, we employ the blinding technology to protect the privacy of the privacy gradients. If no more than $\boldsymbol{n}$ -2 of $\boldsymbol{n}$ participants collude with the aggregation server, VFL could guarantee the encrypted gradients of other participants not being inverted. Experimental evaluations corroborate the practical performance of the presented VFL with high accuracy and efficiency.

Federated learning can combine a large number of scattered user groups and train models collaboratively without uploading data sets, so as to avoid the server collecting user sensitive data. However, the model of federated learning will expose the training set information of users, and the uneven amount of data owned by users in multiple users’ scenarios will lead to the inefficiency of training. In this paper, we propose a privacy-preserving federated learning scheme in fog computing. Acting as a participant, each fog node is enabled to collect Internet of Things (IoT) device data and complete the learning task in our scheme. Such design effectively improves the low training efficiency and model accuracy caused by the uneven distribution of data and the large gap of computing power. We enable IoT device data to satisfy ε-differential privacy to resist data attacks and leverage the combination of blinding and Paillier homomorphic encryption against model attacks, which realize the security aggregation of model parameters. In addition, we formally verified our scheme can not only guarantee both data security and model security, but completely resist collusion attacks launched by multiple malicious entities. Our experiments based on the Fashion-MNIST dataset prove that our scheme is highly efficient in practice.

The maturity of cloud computing, the Internet of Things technology, and intelligent transportation system has promoted the rapid development of vehicular ad-hoc networks (VANETs). To keep pace with real-world demands (mobility, low latency, etc.) in a practical VANETs deployment, there have been attempts to integrate fog computing with VANETs. To facilitate secure interaction in fog-based VANETs, we design a new authenticated key agreement protocol without bilinear pairing. This protocol achieves mutual authentication, generates a securely agreed session key for secret communication, and supports privacy protection. We also give a strict formal security proof and demonstrate how the proposed protocol meets the security requirements in the fog-based VANETs. We then evaluate the efficiency of the proposed protocol, and it shows the practicality of the protocol.

The sharing of electronic health records (EHRs) has great positive significance for research of disease and doctors’ diagnosis. In recent years, cloud-based electronic medical record sharing scheme has brought a lot of conveniences, but the centralization of cloud exposes threats inevitably to data security and privacy preservation. Blockchain technology can be seen as a promising solution to address these problems on account of its unique propertis of decentration, anonymity, unforgeability and verifiability. In this paper, we propose a blockchain based secure and privacy-preserving EHR sharing protocol. Data requester can search desired keyword from data provider to find relevant EHRs on the EHR consortium blockchain and get the re-encryption ciphertext from cloud server after getting the data owner’s authorization. The scheme mainly uses searchable encryption and conditional proxy re-encryption to realize data security, privacy preservation, and access control. Furthermore, proof of authorization is designed as the consensus mechanism for consortium blockchain to guarantee system’s availability. Security analysis demonstrates that the proposed protocol can achieve security goals. Besides, we emulate the cryptographic primitives and implement the proposed scheme on Ethereum platform. Performance evaluation shows that the proposed scheme has high computational efficiency.

The Internet of Things (IoT) devices possessed by individuals produce massive amounts of data. The private data onto specific IoT devices can be combined with intelligent platform to provide help for future research and prediction. As an important digital asset, individuals can sell private data to get rewards. Problems, such as privacy, security, and access control prevent individuals from sharing their private data. The blockchain technology is widely used to build an anonymous trading system. In this article, we construct a blockchain-based privacy-preserving and rewarding private data-sharing scheme (BPRPDS) for IoT. A privacy issue worth considering is that the malicious cloud server may establish a behavior profile database of data users (DUs). In the case of anonymity, the transactions of private data sharing are easy to cause disputes. When anonymous DUs are framed, it is hard to protect their rights. With the help of the deniable ring signature and Monero, we realize the behavior profile building prevention and nonframeability of BPRPDS. At the same time, we utilize the licensing technology executed by smart contracts to ensure flexible access control of multisharing. The proposed BPRPDS is provably secure. Performance analysis and experimental results show that BPRPDS is efficient and practical.