Privacy‐preserving federated learning cyber‐threat detection for intelligent transport systems with blockchain‐based security

Abstract

Artificial intelligence (AI) techniques implemented at a large scale in intelligent transport systems (ITS), have considerably enhanced the vehicles' autonomous behaviour in making independent decisions about cyber threats, attacks, and faults. While, AI techniques are based on data sharing among the vehicles, it is important to note that sensitive data cannot be shared. Thus, federated learning (FL) has been implemented to protect privacy in vehicles. On the other hand, the integrity of data and the safety of aggregation are ensured by using blockchain technology. This paper applied classification approaches to VANET and ITS cyber‐threats detection at the vehicle. Subsequently, by using blockchain and by applying an aggregation strategy to different models, models from the previous step were uploaded in a smart contract. Lastly, we returned the updated models to the vehicles. Furthermore, we conducted an experimental study to measure the effectiveness of the proposed prototype. In this paper, the VeReMi data set was distributed in a balanced manner into five parts in the experimental study. Thus, classification techniques were executed by each vehicle separately, and models were generated. Upon the aggregation of the models in blockchain, they were returned to the vehicles. Lastly, the vehicles updated their decision functions and accessed the precision and accuracy of cyber‐threat detection. The results indicated that the precision and accuracy decreased by 7.1% on average with comparable F1‐score and recall. Our solution ensures the privacy preservation of vehicles whereas blockchain guarantees the safety of aggregation technique and low gas consumption.

Evaluating Federated Learning for intrusion detection in Internet of Things: Review and challenges

“…Indeed, [8] proposes a federated approach based on neural networks to build a multiclass classification approach for detecting the attacks contained in the VeReMi dataset. Furthermore, [38] proposes a federated approach using blockchain where several supervised learning techniques are tested on an extended version of such dataset [39]. In the case of [40], the authors propose an approach based on a semi-supervised model using a neural network and a subset of labeled data for the initial training phase.…”

“…Furthermore, we carry out a pre-processing of the VeReMi dataset to deal with the impact of non-iid data distributions on the effectiveness of the misbehavior detection approach. The main reason is that our approach (unlike other recent works, such as [38]) considers each vehicle as an FL client using its own data during the federated training process. Indeed, as shown in our previous works [22], non-iid data distributions might have a dramatic impact on the effectiveness of the FL-enabled system.…”

Blockchain integration for in-vehicle CAN bus intrusion detection systems with ISO/SAE 21434 compliant reporting

“…It was also proposed for automotive manufacturing traceability systems, allowing companies to trace and document the product’s history and relevant production parameters 36 . More recently, Blockchain was also employed in a federated learning cyber-threat detection approach for intelligent transport systems, where its functionality could ensure the integrity of data and smart contracts can represent the machine learning models 37 . Other authors explore the use of Blockchain for on-demand ride-hailing platforms during the recent pandemics to increase customers’ trust by offering safe cars 38 , or develop an evolutionary game model between the car-hailing platform and the government, proposing the Blockchain as a technical governance measure 39 .…”

Machine learning and internet of things applications in enterprise architectures: Solutions, challenges, and open issues

“…Humans and autonomous devices may cooperate beyond their physical vicinity using digital technology–decision‐making benefits from increased sharing and access to enterprise knowledge‐base. In addition, a surge of digital services, processes, economy, and goods push knowledge‐based decision‐making for digital Enterprise Architectures (EA) and Expert Systems (ES), including consumer devices, social networks, and intelligent transportation (Moulahi et al (2023), Deepa and Prabadevi (2020)). A plethora of physical devices are interconnected in current enterprise architectures through wireless Internet data transfer providing a global communications environment.…”

“…transportation (Moulahi et al (2023), Deepa and Prabadevi (2020)). A plethora of physical devices are interconnected in current enterprise architectures through wireless Internet data transfer providing a global communications environment.…”