A Lightweight Privacy-Preserving CNN Feature Extraction Framework for Mobile Sensing

被引:134
作者
Huang, Kai [1 ,2 ]
Liu, Ximeng [3 ,4 ]
Fu, Shaojing [1 ,2 ]
Guo, Deke [5 ]
Xu, Ming [1 ]
机构
[1] Natl Univ Def Technol, Coll Comp, Changsha 410073, Hunan, Peoples R China
[2] State Key Lab Cryptol, Beijing 1816670, Peoples R China
[3] Fuzhou Univ, Coll Math & Comp Sci, Fuzhou 350108, Fujian, Peoples R China
[4] Fuzhou Univ, Key Lab Informat Secur Network Syst, Fuzhou 350108, Fujian, Peoples R China
[5] Natl Univ Def Technol, Sci & Technol Informat Syst Engn Lab, Changsha 410073, Hunan, Peoples R China
基金
中国国家自然科学基金;
关键词
Cryptography; Feature extraction; Sensors; Neurons; Task analysis; Servers; Encryption; Privacy-preserving; CNN; feature extraction; mobile sensing; MULTIPARTY COMPUTATION; CLOUD; EFFICIENT; SYSTEM;
D O I
10.1109/TDSC.2019.2913362
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
The proliferation of various mobile devices equipped with cameras results in an exponential growth of the amount of images. Recent advances in the deep learning with convolutional neural networks (CNN) have made CNN feature extraction become an effective way to process these images. However, it is still a challenging task to deploy the CNN model on the mobile sensors, which are typically resource-constrained in terms of the storage space, the computing capacity, and the battery life. Although cloud computing has become a popular solution, data security and response latency are always the key issues. Therefore, in this paper, we propose a novel lightweight framework for privacy-preserving CNN feature extraction for mobile sensing based on edge computing. To get the most out of the benefits of CNN with limited physical resources on the mobile sensors, we design a series of secure interaction protocols and utilize two edge servers to collaboratively perform the CNN feature extraction. The proposed scheme allows us to significantly reduce the latency and the overhead of the end devices while preserving privacy. Through theoretical analysis and empirical experiments, we demonstrate the security, effectiveness, and efficiency of our scheme.
引用
收藏
页码:1441 / 1455
页数:15
相关论文
共 59 条
[1]   Deep Learning with Differential Privacy [J].
Abadi, Martin ;
Chu, Andy ;
Goodfellow, Ian ;
McMahan, H. Brendan ;
Mironov, Ilya ;
Talwar, Kunal ;
Zhang, Li .
CCS'16: PROCEEDINGS OF THE 2016 ACM SIGSAC CONFERENCE ON COMPUTER AND COMMUNICATIONS SECURITY, 2016, :308-318
[2]  
[Anonymous], 2014, Comput. Sci.
[3]  
[Anonymous], 2017, ARXIV170508963
[4]   Optimized Honest-Majority MPC for Malicious Adversaries - Breaking the 1 Billion-Gate Per Second Barrier [J].
Araki, Toshinori ;
Barak, Assi ;
Furukawa, Jun ;
Lichter, Tamar ;
Lindell, Yehuda ;
Nof, Ariel ;
Ohara, Kazuma ;
Watzman, Adi ;
Weinstein, Or .
2017 IEEE SYMPOSIUM ON SECURITY AND PRIVACY (SP), 2017, :843-862
[5]   High-Throughput Semi-Honest Secure Three-Party Computation with an Honest Majority [J].
Araki, Toshinori ;
Furukawa, Jun ;
Lindell, Yehuda ;
Nof, Ariel ;
Ohara, Kazuma .
CCS'16: PROCEEDINGS OF THE 2016 ACM SIGSAC CONFERENCE ON COMPUTER AND COMMUNICATIONS SECURITY, 2016, :805-817
[6]  
BEAVER D, 1992, LECT NOTES COMPUT SC, V576, P420
[7]  
Ben-David A, 2008, CCS'08: PROCEEDINGS OF THE 15TH ACM CONFERENCE ON COMPUTER AND COMMUNICATIONS SECURITY, P257
[8]  
Blakley GR, 1979, P AFIPS NEW YORK NY, P313, DOI DOI 10.1109/MARK.1979.8817296
[9]   High-performance secure multi-party computation for data mining applications [J].
Bogdanov, Dan ;
Niitsoo, Margus ;
Toft, Tomas ;
Willemson, Jan .
INTERNATIONAL JOURNAL OF INFORMATION SECURITY, 2012, 11 (06) :403-418
[10]  
Bogdanov D, 2008, LECT NOTES COMPUT SC, V5283, P192