Privacy-preserving convolutional neural network inference scheme based on homomorphic ciphertext transformation
To solve the problems of frequent interaction and low prediction accuracy of existing privacy-protected convolutional neural networks, a homomorphic friendly non-interactive privacy-protected convolutional neural network inference scheme was proposed via homomorphic ciphertext transformation. Utiliz...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Department of Journal on Communications
2024-10-01
|
| Series: | Tongxin xuebao |
| Subjects: | |
| Online Access: | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024216/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841537104795926528 |
|---|---|
| author | LI Ruiqi YI Qin HUANG Yixuan JIA Chunfu |
| author_facet | LI Ruiqi YI Qin HUANG Yixuan JIA Chunfu |
| author_sort | LI Ruiqi |
| collection | DOAJ |
| description | To solve the problems of frequent interaction and low prediction accuracy of existing privacy-protected convolutional neural networks, a homomorphic friendly non-interactive privacy-protected convolutional neural network inference scheme was proposed via homomorphic ciphertext transformation. Utilizing the Pegasus framework, CKKS (Cheon-Kim-Kim-Song) ciphertext was used to parallelize convolution operations in convolution layer. In the activation layer and pooling layer, LWE ciphertext and LUT (look-up table) technology were used to calculate the activation function, maximum pooling and global pooling. Using the ciphertext conversion technology provided by the Pegasus framework, the conversion between different forms of homomorphic ciphertext is realized. Theoretical analysis and experimental results show that the proposed scheme can ensure data security, and has higher inference accuracy and lower calculation and communication overhead. |
| format | Article |
| id | doaj-art-c13aff88201644c19a731914a996a0bc |
| institution | Kabale University |
| issn | 1000-436X |
| language | zho |
| publishDate | 2024-10-01 |
| publisher | Editorial Department of Journal on Communications |
| record_format | Article |
| series | Tongxin xuebao |
| spelling | doaj-art-c13aff88201644c19a731914a996a0bc2025-01-14T08:46:50ZzhoEditorial Department of Journal on CommunicationsTongxin xuebao1000-436X2024-10-0145122379872355Privacy-preserving convolutional neural network inference scheme based on homomorphic ciphertext transformationLI RuiqiYI QinHUANG YixuanJIA ChunfuTo solve the problems of frequent interaction and low prediction accuracy of existing privacy-protected convolutional neural networks, a homomorphic friendly non-interactive privacy-protected convolutional neural network inference scheme was proposed via homomorphic ciphertext transformation. Utilizing the Pegasus framework, CKKS (Cheon-Kim-Kim-Song) ciphertext was used to parallelize convolution operations in convolution layer. In the activation layer and pooling layer, LWE ciphertext and LUT (look-up table) technology were used to calculate the activation function, maximum pooling and global pooling. Using the ciphertext conversion technology provided by the Pegasus framework, the conversion between different forms of homomorphic ciphertext is realized. Theoretical analysis and experimental results show that the proposed scheme can ensure data security, and has higher inference accuracy and lower calculation and communication overhead.http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024216/privacy-preservingconvolutional neural networkhomomorphic encryptionciphertext transformation |
| spellingShingle | LI Ruiqi YI Qin HUANG Yixuan JIA Chunfu Privacy-preserving convolutional neural network inference scheme based on homomorphic ciphertext transformation Tongxin xuebao privacy-preserving convolutional neural network homomorphic encryption ciphertext transformation |
| title | Privacy-preserving convolutional neural network inference scheme based on homomorphic ciphertext transformation |
| title_full | Privacy-preserving convolutional neural network inference scheme based on homomorphic ciphertext transformation |
| title_fullStr | Privacy-preserving convolutional neural network inference scheme based on homomorphic ciphertext transformation |
| title_full_unstemmed | Privacy-preserving convolutional neural network inference scheme based on homomorphic ciphertext transformation |
| title_short | Privacy-preserving convolutional neural network inference scheme based on homomorphic ciphertext transformation |
| title_sort | privacy preserving convolutional neural network inference scheme based on homomorphic ciphertext transformation |
| topic | privacy-preserving convolutional neural network homomorphic encryption ciphertext transformation |
| url | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024216/ |
| work_keys_str_mv | AT liruiqi privacypreservingconvolutionalneuralnetworkinferenceschemebasedonhomomorphicciphertexttransformation AT yiqin privacypreservingconvolutionalneuralnetworkinferenceschemebasedonhomomorphicciphertexttransformation AT huangyixuan privacypreservingconvolutionalneuralnetworkinferenceschemebasedonhomomorphicciphertexttransformation AT jiachunfu privacypreservingconvolutionalneuralnetworkinferenceschemebasedonhomomorphicciphertexttransformation |