When Internet of Consumer Electronics Meets Zero Trust via Zero-Knowledge Proof

Abstract

In the coming 6G era, Internet of Consumer Electronics (ICE) is promising in academia and daily life. To improve the intelligence and reliability of ICE devices with limited resources, a cloud-edge-end collaborative intelligent architecture is designed. However, the current collaborative intelligent ICE system faces multiple security threats, e.g., illegal access to the intelligent data/model resources and poisoning/backdoor attacks during collaborative intelligent model training. The security of the intelligent collaborative ICE is still an open issue. To solve this problem, we propose a zero-knowledge proof (ZKP)-driven zero trust method to protect the security of the intelligent collaborative ICE. First, a zero-trust intelligent collaborative ICE security framework is established for the confidentiality and availability protection of data/model resources. Second, we propose a zero-trust multi-factor access control mechanism for mutual authentication and trust evaluation-based resource access authorization. Third, the ZKP-driven collaborative intelligent ICE mechanism is designed. In this mechanism, the selective differential privacy-based data preservation scheme and the zero-knowledge collaborative intelligent model training scheme are devised. Finally, experimental results demonstrate the effectiveness and efficiency of the proposed secure intelligent collaborative ICE.