The rapid development of Internet of Things (IoT) systems has led to the problem of managing and analyzing the large volumes of data that they generate. Traditional approaches that involve collection of data from IoT devices into one centralized repository for further analysis are not always applicable due to the large amount of collected data, the use of communication channels with limited bandwidth, security and privacy requirements, etc. Federated learning (FL) is an emerging approach that allows one to analyze data directly on data sources and to federate the results of each analysis to yield a result as traditional centralized data processing. FL is being actively developed, and currently, there are several open-source frameworks that implement it. This article presents a comparative review and analysis of the existing open-source FL frameworks, including their applicability in IoT systems. The authors evaluated the following features of the frameworks: ease of use and deployment, development, analysis capabilities, accuracy, and performance. Three different data sets were used in the experiments—two signal data sets of different volumes and one image data set. To model low-power IoT devices, computing nodes with small resources were defined in the testbed. The research results revealed FL frameworks that could be applied in the IoT systems now, but with certain restrictions on their use.

Wearable devices and smartphones that are used to monitor the activity and the state of the driver collect a lot of sensitive data such as audio, video, location and even health data. The analysis and processing of such data require observing the strict legal requirements for personal data security and privacy. The federated learning (FL) computation paradigm has been proposed as a privacy-preserving computational model that allows securing the privacy of the data owner. However, it still has no formal proof of privacy guarantees, and recent research showed that the attacks targeted both the model integrity and privacy of the data owners could be performed at all stages of the FL process. This paper focuses on the analysis of the privacy-preserving techniques adopted for FL and presents a comparative review and analysis of their implementations in the open-source FL frameworks. The authors evaluated their impact on the overall training process in terms of global model accuracy, training time and network traffic generated during the training process in order to assess their applicability to driver’s state and behaviour monitoring. As the usage scenario, the authors considered the case of the driver’s activity monitoring using the data from smartphone sensors. The experiments showed that the current implementation of the privacy-preserving techniques in open-source FL frameworks limits the practical application of FL to cross-silo settings.