Homomorphic Encryption for Sensitive Data on Insecure Edge Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computing devices deployed in physically insecure locations face risks of malicious attacks and malware, which can compromise sensitive information like personally identifiable information (PII) by exposing it in memory during operations.
Innovation Solution
Implementing homomorphic encryption to encrypt sensitive information, allowing operations to be performed on the encrypted data without decrypting it, thereby maintaining security and integrity even if the host edge device is compromised.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensitive information is stored and processed in plaintext on edge devices, then operations can be performed efficiently, but security is compromised when devices are located in physically insecure locations
Solution Approach 1:
The system segments data into sensitive and non-sensitive categories, applying different handling mechanisms. Sensitive data is encrypted while non-sensitive data remains accessible, allowing selective protection without compromising overall system operation. This segmentation enables the system to maintain security for critical information while preserving ease of operation for other data.
Solution Approach 2:
Homomorphic encryption acts as an intermediary mechanism that enables operations on encrypted data without requiring decryption. This intermediary layer allows the system to maintain security (encrypted data in memory) while still enabling data processing operations, effectively mediating between the conflicting requirements of security and operational accessibility.
2Reliability
If data is encrypted using traditional methods, then security is improved, but operations require decryption which exposes plaintext in memory
Solution Approach 1:
The system changes the cryptographic parameter from traditional symmetric encryption to homomorphic encryption. This parameter change fundamentally alters how encryption works, allowing mathematical operations to be performed directly on ciphertext without decryption. The result is that sensitive data remains encrypted in memory during operations, eliminating plaintext exposure while maintaining security.
3Object-affected harmful factors
If homomorphic encryption is implemented, then plaintext exposure is prevented, but computational overhead increases
Solution Approach 1:
The system applies homomorphic encryption selectively rather than universally. By identifying and encrypting only the sensitive portions of data that require protection, the system reduces the overall computational overhead compared to encrypting all data. This partial application approach maintains security for critical information while minimizing the energy and computational cost associated with homomorphic operations.
Data Source
AI summary
Methods and systems for managing data in a system comprising edge devices and a data transfer bus is disclosed. The data may be managed by encrypting with homomorphic encryption and storing the data in edge devices. The edge devices may be stored in a physically insecure locations and maintain encryption of the data between edge devices. The data that is stored and in an encrypted state may include personally identifiable information. Maintaining encryption of the personally identifiable information between edge devices may prevent exposure to memory of the edge devices. Preventing exposure to memory of the personally identifiable identification may prevent a likelihood of a malicious attack and/or malware from finding the personally identifiable information if the malicious attack and/or malware may monitor memory of the edge devices.