A Secure and Privacy-Preserving Machine Learning Model Sharing Scheme for Edge-Enabled IoT

Meng Han,Hao Xu,Xianfei Zhou,Naiyu Wang,Kai Xu,Jianlin Jiao,Ning Dong

doi:10.1109/access.2021.3051945

Abstract

With the popular use of IoT devices, edge computing has been widely applied in the Internet of things (IoT) and regarded as a promising solution for its wide distribution, decentralization, low latency. At the same time, in response to the massive computing data and intelligent requirements of various applications in the IoT, artificial intelligence (AI) technology has also achieved rapid development. As a result, edge intelligence (EI) for the Internet of Things has attracted widespread attention. Driven by the requirement that making full use of data, machine learning (ML) models trained in EI are usually shared. However, there may be some security and privacy issues due to the openness and heterogeneity of edge intelligence. How to ensure flexible data access and data security as well as the accountability for edge nodes and users in EI model sharing have become important issues. In this article, we propose a Ciphertext Policy Attribute Based Proxy Re-encryption (CP-ABPRE) scheme with accountability to address the security and privacy issues in EI model sharing. In our scheme, a user can delegate the access right to others to make model access more flexible. Furthermore, each entity that may need to be held accountable is embedded a unique ID to achieve traceability. Finally, security analysis and performance evaluation are given to prove that our scheme is CPA secure and does not lose much efficiency with more features.

Highlights

Cisco’s report [1] has predicted that in the future, a lot of Internet of things (IoT) data will be generated from the edge side
If these huge data are processed by cloud computing, the process of sending them to the cloud will consume a lot of bandwidth resource and bring great computing pressure to the cloud [2]
Users can judge the behavior of the edge nodes in our scheme using a reasonable accountability checking mechanism

Summary

INTRODUCTION

Cisco’s report [1] has predicted that in the future, a lot of IoT data will be generated from the edge side. X. Zhou et al.: Secure and Privacy-Preserving ML Model Sharing Scheme for Edge-Enabled IoT the model, encrypt the data under the access policy. 3) We set Alice and Bob as the model requesters, they can get the requested ciphertext from the edge node, and if their attributes satisfy the access structure, the CT can be decrypted successfully. 4) If Alice wants to delegate her access right to Bob who does not have access to the corresponding model parameters, she first generates a re-encryption key (RK) according to her own secret key and a specified access structure AS∗ and sends it to the proxy server (the edge node). Re_Enc(IDct , CT, PK, RK)→CTRE or ⊥ This algorithm is responsible for generating a re-encrypted ciphertext CTRE under a new access structure. Definition 2: Our CP-ABPRE scheme is chosen plaintext attack (CPA) secure if no Probabilistic Polynomial Time (PPT) adversary A can win the following game with non-negligible advantage

DESCRIPTION OF OUR SYSTEM

ACCOUNTABILITY

SECURITY ANALYSIS

Findings

CONCLUSION