Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Two Birds with One Stone: Differential Privacy by Low-power SRAM Memory

Jianqing Liu, Na Gong, Hritom Das

TL;DR

The design realizes local differential privacy (LDP) by harnessing the inherent hardware noise into controlled LDP noise when data is stored in the memory, which leads to double-faceted gains in privacy and power efficiency.

Abstract

The software-based implementation of differential privacy mechanisms has been shown to be neither friendly for lightweight devices nor secure against side-channel attacks. In this work, we aim to develop a hardware-based technique to achieve differential privacy by design. In contrary to the conventional software-based noise generation and injection process, our design realizes local differential privacy (LDP) by harnessing the inherent hardware noise into controlled LDP noise when data is stored in the memory. Specifically, the noise is tamed through a novel memory design and power downscaling technique, which leads to double-faceted gains in privacy and power efficiency. A well-round study that consists of theoretical design and analysis and chip implementation and experiments is presented. The results confirm that the developed technique is differentially private, saves 88.58% system power, speeds up software-based DP mechanisms by more than 10^6 times, while only incurring 2.46% chip overhead and 7.81% estimation errors in data recovery.

Two Birds with One Stone: Differential Privacy by Low-power SRAM Memory

TL;DR

The design realizes local differential privacy (LDP) by harnessing the inherent hardware noise into controlled LDP noise when data is stored in the memory, which leads to double-faceted gains in privacy and power efficiency.

Abstract

The software-based implementation of differential privacy mechanisms has been shown to be neither friendly for lightweight devices nor secure against side-channel attacks. In this work, we aim to develop a hardware-based technique to achieve differential privacy by design. In contrary to the conventional software-based noise generation and injection process, our design realizes local differential privacy (LDP) by harnessing the inherent hardware noise into controlled LDP noise when data is stored in the memory. Specifically, the noise is tamed through a novel memory design and power downscaling technique, which leads to double-faceted gains in privacy and power efficiency. A well-round study that consists of theoretical design and analysis and chip implementation and experiments is presented. The results confirm that the developed technique is differentially private, saves 88.58% system power, speeds up software-based DP mechanisms by more than 10^6 times, while only incurring 2.46% chip overhead and 7.81% estimation errors in data recovery.
Paper Structure (24 sections, 4 theorems, 16 equations, 17 figures, 3 tables)

This paper contains 24 sections, 4 theorems, 16 equations, 17 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Local Differential Privacy
  4. SRAM Failure Characteristics
  5. Privacy by Design in SRAM Memory
  6. Principle of SRAM_DP
  7. Hardware Architecture of SRAM_DP
  8. Privacy and Utility Analysis
  9. Compliance to LDP Notion
  10. Side-channel Analysis
  11. Utility Analysis
  12. Statistics Recovery Algorithms
  13. EM Algorithm
  14. CLR Algorithm
  15. Performance Evaluation
  16. ...and 9 more sections

Key Result

Theorem 4.1

The SRAM_DP mechanism satisfies $\epsilon_{\infty}$-differential privacy for any input bit strings $X$, where $\epsilon_{\infty}$ = $\sum_{i=1}^{n} \text{ln}(\frac{1 - \frac{1}{2}f_{i}}{\frac{1}{2}f_{i}})$.

Figures (17)

  • Figure 1: 45nm SRAM cell schematics with the smallest silicon area: (a) 6T (C61) and (b) 8T (C81).
  • Figure 2: Failure characteristics of 45 nm 6T and 8T cells.
  • Figure 3: A toy example showing the procedures of SRAM_DP.
  • Figure 4: Hardware Architecture of the proposed SRAM_DP.
  • Figure 5: Privacy analysis.
  • ...and 12 more figures

Theorems & Definitions (9)

  • Definition 2.1: Local Differential Privacy erlingsson2014rappor
  • Definition 2.2: Randomized Response warner1965randomized
  • Theorem 4.1
  • proof
  • Theorem 4.2
  • proof
  • Lemma 4.3
  • Theorem 7.1
  • proof