Abstract
Federated learning (FL) has been gaining popularity as a way to provide privacy-preserving data sharing for the Internet of Medical Things (IoMT). As a complementary, blockchain technology is used in recent literature to make FL secure. However, existing blockchain-based FL (BFL) solutions do not perform well when data in a BFL cluster are sparse. A direct solution is to collect as many devices as possible to establish a large BFL cluster. However, these devices may locate in geographically distant areas and be separated by great distance, which further results in high communication latency. The high latency will lead to BFL’s low system efficiency due to frequent communications in the blockchain consensus. In this article, we propose that the large cluster should be divided into multiple smaller clusters, each in its own geographical area and organized with a BFL. In this context, we propose CFL, a cross-cluster FL system facilitated by the cross-chain technique. CFL connects multiple BFL clusters, where only a few aggregated updates are transmitted over long distances across clusters, thus improving the system efficiency. The design of CFL focuses on a cross-chain consensus protocol, which guarantees the model updates to be exchanged securely across clusters. We carry out extensive experiments to evaluate CFL in comparison with BFL, and show both CFL’s feasibility and efficiency.
Highlights
N OWADAYS, a growing number of medical devices are connected to build a new network, namely, Internet of Medical Things (IoMT) [1]
By aggregating the data generated in different devices, IoMT is expected to contribute to a valuable machine learning (ML) model, which can be useful in multiple scenarios, such as health monitoring, auxiliary diagnoses, and pathophoresis prediction [2]
Corresponding to different consensus mechanisms designed for CFL, two prototypes are implemented, respectively, namely, CFL-hasty consensus (HstCon) and CFL-deferred consensus (DefCon)
Summary
N OWADAYS, a growing number of medical devices are connected to build a new network, namely, Internet of Medical Things (IoMT) [1]. If the number of IoMT devices in the hospital is small, it would be too long for the BFL cluster to accumulate enough data. To deal with the data sparsity and privacy leakage problems while providing high system efficiency, we propose a cross-cluster FL framework via cross-chain technique (CFL) in this article. BFL is conducted with the model updates being aggregated. JIN et al.: CROSS-CLUSTER FEDERATED LEARNING AND BLOCKCHAIN FOR INTERNET OF MEDICAL THINGS few aggregated updates are transmitted over a long distance in CFL. To enable the secure cross-cluster model exchange, we design two consensus mechanisms of the blockchain, namely, hasty consensus (HstCon) and deferred consensus (DefCon). 1) We identify the difficulty in the existing BFL solutions, namely, the problem of data sparsity and the problem of low efficiency plus privacy leakage. Our major contributions include the following. 1) We identify the difficulty in the existing BFL solutions, namely, the problem of data sparsity and the problem of low efficiency plus privacy leakage. 2) We propose CFL to deal with the difficulty by extending the single-cluster FL to cross-cluster FL, which takes advantage of cross-chain technology to provide secure communications across clusters. 3) Prototypes of CFL are implemented and extensive experiments are conducted to demonstrate its feasibility and efficiency
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
