Towards edge-collaborative, lightweight and privacy-preserving classification framework
Aiming at the problems of data leakage of perceptual image and computational inefficiency of privacy-preserving classification framework in edge-side computing environment, a lightweight and privacy-preserving classification framework (PPCF) was proposed to supports encryption feature extraction and...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Department of Journal on Communications
2022-01-01
|
| Series: | Tongxin xuebao |
| Subjects: | |
| Online Access: | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2022004/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841540000915652608 |
|---|---|
| author | Jinbo XIONG Yongjie ZHOU Renwan BI Liang WAN Youliang TIAN |
| author_facet | Jinbo XIONG Yongjie ZHOU Renwan BI Liang WAN Youliang TIAN |
| author_sort | Jinbo XIONG |
| collection | DOAJ |
| description | Aiming at the problems of data leakage of perceptual image and computational inefficiency of privacy-preserving classification framework in edge-side computing environment, a lightweight and privacy-preserving classification framework (PPCF) was proposed to supports encryption feature extraction and classification, and achieve the goal of data transmission and computing security under the collaborative classification process of edge nodes.Firstly, a series of secure computing protocols were designed based on additive secret sharing.Furthermore, two non-collusive edge servers were used to perform secure convolution, secure batch normalization, secure activation, secure pooling and other deep neural network computing layers to realize PPCF.Theoretical and security analysis indicate that PPCF has excellent accuracy and proved to be security.Actual performance evaluation show that PPCF can achieve the same classification accuracy as plaintext environment.At the same time, compared with homomorphic encryption and multi-round iterative calculation schemes, PPCF has obvious advantages in terms of computational cost and communication overhead. |
| format | Article |
| id | doaj-art-e91d2839e3ff4c149680789da61cb99f |
| institution | Kabale University |
| issn | 1000-436X |
| language | zho |
| publishDate | 2022-01-01 |
| publisher | Editorial Department of Journal on Communications |
| record_format | Article |
| series | Tongxin xuebao |
| spelling | doaj-art-e91d2839e3ff4c149680789da61cb99f2025-01-14T06:30:29ZzhoEditorial Department of Journal on CommunicationsTongxin xuebao1000-436X2022-01-014312713759398470Towards edge-collaborative, lightweight and privacy-preserving classification frameworkJinbo XIONGYongjie ZHOURenwan BILiang WANYouliang TIANAiming at the problems of data leakage of perceptual image and computational inefficiency of privacy-preserving classification framework in edge-side computing environment, a lightweight and privacy-preserving classification framework (PPCF) was proposed to supports encryption feature extraction and classification, and achieve the goal of data transmission and computing security under the collaborative classification process of edge nodes.Firstly, a series of secure computing protocols were designed based on additive secret sharing.Furthermore, two non-collusive edge servers were used to perform secure convolution, secure batch normalization, secure activation, secure pooling and other deep neural network computing layers to realize PPCF.Theoretical and security analysis indicate that PPCF has excellent accuracy and proved to be security.Actual performance evaluation show that PPCF can achieve the same classification accuracy as plaintext environment.At the same time, compared with homomorphic encryption and multi-round iterative calculation schemes, PPCF has obvious advantages in terms of computational cost and communication overhead.http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2022004/edge-collaborativeprivacy-preserving object classificationadditive secret sharingdeep neural networkse-cure computing protocol |
| spellingShingle | Jinbo XIONG Yongjie ZHOU Renwan BI Liang WAN Youliang TIAN Towards edge-collaborative, lightweight and privacy-preserving classification framework Tongxin xuebao edge-collaborative privacy-preserving object classification additive secret sharing deep neural network se-cure computing protocol |
| title | Towards edge-collaborative, lightweight and privacy-preserving classification framework |
| title_full | Towards edge-collaborative, lightweight and privacy-preserving classification framework |
| title_fullStr | Towards edge-collaborative, lightweight and privacy-preserving classification framework |
| title_full_unstemmed | Towards edge-collaborative, lightweight and privacy-preserving classification framework |
| title_short | Towards edge-collaborative, lightweight and privacy-preserving classification framework |
| title_sort | towards edge collaborative lightweight and privacy preserving classification framework |
| topic | edge-collaborative privacy-preserving object classification additive secret sharing deep neural network se-cure computing protocol |
| url | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2022004/ |
| work_keys_str_mv | AT jinboxiong towardsedgecollaborativelightweightandprivacypreservingclassificationframework AT yongjiezhou towardsedgecollaborativelightweightandprivacypreservingclassificationframework AT renwanbi towardsedgecollaborativelightweightandprivacypreservingclassificationframework AT liangwan towardsedgecollaborativelightweightandprivacypreservingclassificationframework AT youliangtian towardsedgecollaborativelightweightandprivacypreservingclassificationframework |