Hierarchical Multiparty Homomorphic Encryption for Federated Learning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing multiparty homomorphic encryption protocols face issues such as single points of failure, high time overheads, increased communication costs, and the need for system resets due to configuration updates, especially in federated learning scenarios involving geographically distributed devices like IoT devices.

Innovation Solution

A hierarchical system is implemented with abstracted devices forming clusters, allowing different multiparty homomorphic encryption protocols to be used based on device reliability, reducing communication costs and simplifying configuration updates by isolating clusters and using gateway devices for intermediate aggregation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If fully-multiparty homomorphic encryption (FMHE) is implemented to eliminate single points of failure, then security is improved, but the system becomes sensitive to contributor delays and requires system resets for configuration updates

Engineering Contradiction:
ImprovesecurityVSAvoidtime overhead
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system segments contributors into multiple clusters, each running independent MHE protocols. This segmentation allows the system to avoid full system resets by isolating configuration changes to specific clusters, thereby reducing the time overhead associated with system-wide resets while maintaining security through distributed architecture.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system dynamically selects which MHE protocol each cluster implements based on contributor reliability. More reliable devices run FMHE for higher security, while less reliable devices run PHE. This dynamic allocation optimizes the balance between security and time overhead by matching protocol complexity to device capability.

Inventive Principle:
Principle #15Dynamics

2Reliability

If threshold-multiparty homomorphic encryption (TMHE) is used to improve resilience to dropout, then reliability is improved, but communication costs and time overhead increase

Engineering Contradiction:
Improveresilience to dropoutVSAvoidcommunication costs
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

By segmenting the system into clusters that independently run MHE protocols, communication is localized within clusters rather than requiring system-wide communication. This reduces overall communication costs while maintaining resilience through the threshold-based security model at the cluster level.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different clusters can implement different MHE protocols based on their specific reliability requirements and communication constraints. This allows the system to optimize the balance between resilience and communication costs locally at each cluster rather than applying a uniform approach system-wide.

Inventive Principle:
Principle #3Local quality

3Reliability

If a uniform MHE protocol is implemented across all devices, then security consistency is improved, but flexibility to accommodate varying device reliability is reduced

Engineering Contradiction:
Improvesecurity consistencyVSAvoidflexibility
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system applies different MHE protocols to different clusters based on their specific reliability characteristics. This allows each cluster to have optimized security consistency appropriate to its devices, while the overall system maintains flexibility in protocol selection.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The gateway device serves multiple functions: it acts as a client device participating in cluster MHE operations, and simultaneously serves as a server for other client devices. This multi-functionality enables flexible protocol selection while maintaining security consistency through the gateway's coordination role.

Inventive Principle:
Principle #6Universality (Multi-functionality)

4Adaptability or versatility

If configuration updates are required to add, remove or update contributors, then adaptability is improved, but the entire system must be reset causing loss of time

Engineering Contradiction:
Improveconfiguration flexibilityVSAvoidsystem reset time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system segments contributors into independent clusters, allowing configuration updates to be applied locally to affected clusters rather than requiring system-wide resets. This maintains adaptability for configuration changes while significantly reducing the time overhead by isolating updates to specific clusters only.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The gateway device acts as an intermediary that manages configuration updates within its cluster. When configuration changes are needed, the gateway coordinates the update locally without triggering system-wide resets, thereby maintaining adaptability while reducing time overhead through localized update management.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4697640A1Hierarchical system for secure multiparty computation and associated methods
Publication Date: 2026.02.18 NAGRAVISION SRL
  • EP4697640A1 patent drawingFigure 1
  • EP4697640A1 patent drawingFigure 2
  • EP4697640A1 patent drawingFigure 3

AI summary

A hierarchical system configured to implement a secure multiparty computation protocol comprises: a plurality of client devices arranged in a plurality of C clusters of client devices, each cluster comprising: one or more client devices of the plurality of client devices; and a gateway device configured to communicate with the one or more client devices in the cluster, wherein, within each cluster, the one or more client devices are configured to implement a respective multiparty homomorphic encryption (MHE) protocol associated with that cluster; wherein: the plurality of C gateway devices form a gateway cluster of gateway devices, each gateway device of the plurality of C gateway devices configured to communicate with the other gateway devices; the gateway devices in the gateway cluster are configured to implement a gateway multiparty homomorphic encryption (MHE) protocol; and one or more gateway devices of the plurality of C gateway devices is configured to communicate with a central server.