SRAM Local Differential Privacy via Noise Injection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The broad collection of data from various sources poses risks to individual privacy and data security, as it can be vulnerable to malicious analysis, theft, and unauthorized access.

Innovation Solution

The implementation of local differential privacy (LDP) techniques using static random-access memory (SRAM), which involves key-based shuffling and noise injection to protect data privacy by obscuring sensitive information.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If data is collected from multiple sources to enhance analysis quality, then data diversity and volume increase, but privacy risks and security vulnerabilities worsen

Engineering Contradiction:

Improvedata volumeVSAvoidprivacy risk

Core Design Contradiction:

Quantity of substanceVSObject-affected harmful factors

Solution Approach 1:

The patent converts the harmful effect of data collection into beneficial privacy protection by using the collected data itself to generate noise that masks sensitive information. The data volume that could increase privacy risk is instead used to create statistical noise that protects individual privacy while maintaining analysis utility.

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

Solution Approach 2:

The patent introduces noise as an intermediary element between the raw data and the analysis process. This noise acts as a mediator that obscures the direct link between individual data points and their sources, allowing data to be analyzed without compromising privacy.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Object-affected harmful factors

If noise is added to data to protect privacy, then privacy protection improves, but data utility for analysis deteriorates

Engineering Contradiction:

Improveprivacy protectionVSAvoiddata utility

Core Design Contradiction:

Object-affected harmful factorsVSLoss of information

Solution Approach 1:

The patent changes the parameters of the noise addition process by using data-driven noise generation where the noise characteristics are adapted to the specific data distribution. This allows optimization of the balance between privacy protection and data utility, ensuring that noise is added just enough to protect privacy while preserving analytical value.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent implements feedback mechanisms where the noise generation process is informed by analysis of the data itself. The system continuously adjusts noise parameters based on data characteristics, ensuring that privacy protection is achieved with minimal impact on data utility for downstream analysis tasks.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20250077705A1Local differential privacy using static random-access memory

Publication Date: 2025.03.06 UNIV OF SOUTH ALABAMA

AI summary

Techniques for data manipulation based on local differential privacy using static random-access memory are disclosed. Data that requires differential privacy data manipulation is accessed for storage. The data is prepared for storage using key-based shuffling. Keys are selected by a random key index generator. The shuffled data is stored in a static random-access memory (SRAM). The SRAM comprises high-reliability and low-reliability storage cells. A supply voltage of the SRAM is lowered. The lowering a supply voltage produces a pre-characterized noise level in the SRAM storage cells. Cells among the low-reliability cells are flipped based on the pre-characterized noise level. The flipping cells are used to inject random noise, and their values are provided by random noise generators. A read of the stored data is performed. The read occurs across both high-reliability and low-reliability storage cells. The data that was read is unshuffled, using key-based unshuffling.