Federated learning, as a form of distributed learning, aims to protect the local data while utilizing distributed data to train a global model. However, federated learning still faces challenges related to privacy leakage in Internet of Things (IoT). Researches indicate that the server can infer private information from the local gradients. Additionally, malicious participants may upload poisoning local models, which contaminate the global model and cause a decline in accuracy. Furthermore, irregular participants with low-quality data in the real world can also impact the performance of the global model. Simultaneously addressing these three issues poses a significant challenge. This is because privacy protection strategies in FL are designed to prevent access to the local gradients to avoid information leakage. However, strategies with Byzantine robustness and defense against irregular participants typically require access to the local gradients to calculate the reliability of each participant. Therefore, we use secret sharing as the underlying technology to propose a 3PC privacy-preserving federated learning framework BPFL that can resist Byzantine attacks and irregular participants. Compared with the previous schemes, our scheme can not only protect data privacy but also minimize the negative impact of malicious or irregular participants on the global model. We implemented BPFL and compared it with Mkrum and PPFL. Experimental results indicate that our approach maintains high performance when facing malicious attackers and irregular participants.

Federated learning has shown great potential in Internet of Things (IoTs) for performing intelligent decision making. It allows IoT devices to collaboratively train a neural network upon the data they collect while separately keeping these data staying local. However, several research works have shown that such architecture still faces security challenges that adversaries could raise inference attack to the transferring model parameters to reveal data from devices. Moreover, another security risk in federated learning is that malicious devices may launch model pollution attack to reduce the quality of the aggregated model, or dishonest server may output incorrect aggregated result to the devices. Most existing privacy-preserving federated learning protocols could not deal with both problems. In this paper, we present WVFL, a secure weighted aggregation protocol in which aims to minimize the effect of wrong local models to the aggregated model, meanwhile allowing devices to verify the correctness of the aggregation result. All important intermediate values in the process are in encrypted form so that they would not be revealed to both devices and servers to guarantee privacy. At the end of this paper, we give implementation of our WVFL scheme, showing its efficiency compared with previous work.

Cloud-based Internet of Things (IoT) system is becoming a promising architecture in our modern society. However, cloud-based IoT system brings a number of challenges in the security aspect while improving the efficiency of data analytics. Especially, searching on encrypted data is challenging given today’s technology. Thus searchable encryption has emerged as one of the important research fields. Public key encryption with equality test (PKEET) provides a simple but useful mechanism to cryptographically protect data while keeping it available for equality test on ciphertexts. However, PKEET schemes in the literature are not suitable for cloud-based IoT system with privacy protection enhancement since the untrustworthy cloud server may be interested in query results by itself and hence reveal the private information of data owner out of the expectation of data user. Even worse, it could launch offline message recovery attack (OMRA) based on the returned query results. In this paper, we introduce a new notion of public key encryption with outsourced equality test (PKE-OET) for adapting to cloud-based IoT environments as well as providing a flexible solution to resist against OMRA without taking any information about entrusted parties as input in encryption. We formally define the security model of PKE-OET against three types of adversary including IND-CCA-I, IND-CCA-II and IND-CCA-III. We present a generic PKE-OET construction using a new variant of smooth projective hash function (SPHF) with a novel Lin-Hom property, which is of independent interest. Then we provide an efficient PKE-OET instantiation from Symmetric eXternal Diffie-Helllman (SXDH) assumption and show its practicality for cloud-based IoT environments through a series of experiments on Cloud Server and Raspberry Pi.

Abstract ID-based encryption with equality test (IBEET) allows a tester to compare ciphertexts encrypted under different public keys for checking whether they contain the same message. In this paper, we first introduce group mechanism into IBEET and propose a new primitive, namely group ID-based encryption with equality test (G-IBEET). With the group mechanism: (1) group administrator can authorize a tester to make comparison between ciphertexts of group users, but it cannot compare their ciphertexts with any ciphertext of any user who is not in the group. Such group granularity authorization can make IBEET that adapts to group scenario; (2) for the group granularity authorization, only one trapdoor, named group trapdoor, should be issued to the tester, which can greatly reduce the cost of computation, transmission and storage of trapdoors in traditional IBEET schemes; (3) G-IBEET can resist the insider attack launched by the authorized tester, which is an open problem in IBEET. We give definitions for G-IBEET and propose a concrete construction with an efficient test algorithm. We then give its security analysis in the random oracle model.