Homomorphic Encryption Execution Unit for Secure Cloud Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for executing encrypted machine code do not ensure that data and program components remain encrypted on the target machine, making them vulnerable to unauthorized access, especially in cloud computing scenarios where data is processed by third-party service providers.

Innovation Solution

The method involves executing bitwise homomorphically encrypted machine code using an execution unit that performs Boolean logic operations within an encryption space, where arithmetic functions equivalent to Boolean logic operations are used, ensuring that all calculations remain encrypted and invisible to the target machine.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional encryption methods are used to protect data and programs during transmission, then data security during transmission is improved, but data and program components become unencrypted during processing, making them visible to the resource operator

Engineering Contradiction:
Improvedata securityVSAvoidunauthorized access to data during processing
Core Design Contradiction:
ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The patent changes the fundamental parameter of encryption from binary (encrypted/unencrypted) to a continuous state by implementing homomorphic encryption. This allows data to maintain its encrypted state throughout the entire processing lifecycle, eliminating the vulnerability window where data is unencrypted during execution. The encryption scheme is transformed to support arithmetic operations directly on ciphertexts.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent replaces the traditional mechanical decryption-execution-encryption process with a homomorphic encryption system where arithmetic operations are performed directly on encrypted data. Instead of decrypting data to process it and then re-encrypting the result, the system substitutes the entire decryption-execution cycle with homomorphic arithmetic operations that work natively on ciphertexts.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Ease of manufacture

If data is outsourced to third-party cloud providers for processing, then infrastructure costs and maintenance burden are reduced, but control and visibility over data processing location and security are lost

Engineering Contradiction:
Improveinfrastructure maintenanceVSAvoiddata control and security verification
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

The patent introduces homomorphic encryption as an intermediary layer between the client and cloud provider. This intermediary mechanism allows the client to outsource processing to third parties while maintaining control and security verification. The encryption scheme acts as a trusted mediator that enables the cloud provider to process data without ever seeing the plaintext, thus maintaining security and control even in distributed cloud environments.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent segments the data processing function into separate components: the client retains control of the encryption keys and sensitive data, while the cloud provider performs only the encrypted arithmetic operations. This segmentation allows outsourcing of infrastructure while maintaining security control, as the cloud provider's role is limited to executing homomorphic operations without access to the actual data or keys.

Inventive Principle:
Principle #1Segmentation

3Productivity

If machine code is decrypted for execution on the target machine, then processing speed and performance are improved, but the data and program components become visible and vulnerable to unauthorized access

Engineering Contradiction:
Improveprocessing speedVSAvoidunauthorized viewing of data
Core Design Contradiction:
ProductivityVSObject-generated harmful factors

Solution Approach 1:

The patent fundamentally changes the parameter of data state during processing from decrypted/encrypted binary states to a homomorphic encrypted state that supports arithmetic operations. This allows the system to maintain high processing speed by performing native arithmetic operations on ciphertexts directly, eliminating the need to switch between encrypted and decrypted states, thus maintaining both speed and security simultaneously.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2678772B1Encrypted computing
Publication Date: 2015.04.08 GOTTFRIED WILHELM LEIBNIZ UNIV HANNOVER
  • EP2678772B1 patent drawingFigure 1
  • EP2678772B1 patent drawingFigure 2
  • EP2678772B1 patent drawingFigure 3

AI summary

The invention relates to a method for executing a bit-by-bit homomorphically encrypted machine code by means of an execution unit arranged on a data processing system, wherein the encrypted machine code contains instructions which can be executed by the execution unit, comprising the provision of arithmetic functions in which the operands and the ensuing result correspond to a Boolean logic function, and the execution of the instructions contained in the encrypted machine code by the execution unit using Boolean logic operations, the computation of which is carried out by the arithmetic function corresponding to the respective Boolean logic operation on the basis of the encrypted bits in the encryption space.