Guaranteeing Local Differential Privacy on Ultra-low-power Systems

被引:31
作者
Choi, Woo-Seok [1 ]
Tomei, Matthew [1 ]
Vicarte, Jose Rodrigo Sanchez [1 ]
Hanumolu, Pavan Kumar [1 ]
Kumar, Rakesh [1 ]
机构
[1] Univ Illinois, Urbana, IL 61801 USA
来源
2018 ACM/IEEE 45TH ANNUAL INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE (ISCA) | 2018年
关键词
differential privacy; randomized response; RAPPOR; microcontrollers; low-power systems; IoT; DISTRIBUTIONS;
D O I
10.1109/ISCA.2018.00053
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Sensors in mobile devices and IoT systems increasingly generate data that may contain private information of individuals. Generally, users of such systems are willing to share their data for public and personal benefit as long as their private information is not revealed. A fundamental challenge lies in designing systems and data processing techniques for obtaining meaningful information from sensor data, while maintaining the privacy of the data and individuals. In this work, we explore the feasibility of providing local differential privacy on ultra-low-power systems that power many sensor and IoT applications. We show that low resolution and fixed point nature of ultra-low-power implementations prevent privacy guarantees from being provided due to low quality noising. We present techniques, resampling and thresholding, to overcome this limitation. The techniques, along with a privacy budget control algorithm, are implemented in hardware to provide privacy guarantees with high integrity. We show that our hardware implementation, DP-Box, has low overhead and provides high utility, while guaranteeing local differential privacy, for a range of sensor/IoT benchmarks.
引用
收藏
页码:561 / 574
页数:14
相关论文
共 40 条
[1]  
[Anonymous], IDENTIFYING QUALITAT
[2]  
[Anonymous], I2C BUS SPEC US MAN
[3]  
[Anonymous], CLOUD 2009 ICSE
[4]  
[Anonymous], IEEE COMMUNICATIONS
[5]  
[Anonymous], TRANSPORT RES C EMER
[6]  
Chakraborty S., 2013, Proceedings of the 14th Workshop on Mobile Computing Systems and Applications, P11
[7]   Distributed data mining in credit card fraud detection [J].
Chan, PK ;
Fan, W ;
Prodromidis, AL ;
Stolfo, SJ .
IEEE INTELLIGENT SYSTEMS & THEIR APPLICATIONS, 1999, 14 (06) :67-74
[8]  
Chaudhuri Kamalika, 2011, JMLR Workshop Conf Proc, V2011, P155
[9]   Hardware generation of arbitrary random number distributions from uniform distributions via the inversion method [J].
Cheung, Ray C. C. ;
Lee, Dong-U ;
Luk, Wayne ;
Villasenor, John D. .
IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS, 2007, 15 (08) :952-962
[10]   Uniqueness of medical data mining [J].
Cios, KJ ;
Moore, GW .
ARTIFICIAL INTELLIGENCE IN MEDICINE, 2002, 26 (1-2) :1-24