Abstract
With the development of the Internet of Things (IoT), the massive data sharing between IoT devices improves the Quality of Service (QoS) and user experience in various IoT applications. However, data sharing may cause serious privacy leakages to data providers. To address this problem, in this study, data sharing is realized through model sharing, based on which a secure data sharing mechanism, called BP2P-FL, is proposed using peer-to-peer federated learning with the privacy protection of data providers. In addition, by introducing the blockchain to the data sharing, every training process is recorded to ensure that data providers offer high-quality data. For further privacy protection, the differential privacy technology is used to disturb the global data sharing model. The experimental results show that BP2P-FL has high accuracy and feasibility in the data sharing of various IoT applications.
1. Introduction
With the development of the Internet technology, the Internet of Things (IoT) is widely used in various industries [1]. Sensors are an important part of the IoT and the most important data source for the IoT system. The perception data collected by a single sensor often cannot meet users needs, and the real value of the IoT lies in the comprehensive utilization and sharing of various data and information [2, 3, 4]. For example, in healthcare, data sharing can provide valuable health records, including treatment and physical examination information, and can offer more targeted treatments for patients. In industry, by analyzing the collected data, data sharing can accurately understand the preferences of tourists and predict future tourism hot spots to improve the quality of service. However, data sharing in IoT may face various problems. First, it is very difficult for each pair of organizations to build mutual trust. As a result, it is unlikely to share reliable local data. Second, data privacy has become a big problem that hinders data sharing because data owners suffer from privacy leakage. Therefore, achieving effective data sharing is a challenge, especially when these two problems have not been solved.
