Homomorphic Encryption for Secure Federated Learning: A PRISMA-Based Systematic Review
Abstract
Federated learning (FL) enables collaborative training without sharing raw data but remains vulnerable to privacy leakage, poisoned gradients, and incorrect aggregation. Homomorphic encryption (HE) supports secure aggregation over encrypted updates, yet existing solutions are heterogeneous and lack systematic comparison. This paper presents a PRISMA-based systematic review of 104 studies on HE-enabled security mechanisms in FL, analyzed within the confidentiality–integrity–availability (CIA) framework. The results show a growing focus on poisoned-gradient robustness and verifiable aggregation, while only few integrated approaches address all CIA dimensions. These findings outline current limitations and guide future development of secure and trustworthy federated learning systems.
How strong is the evidence?
Evidence Hub reads this as a systematic review. Its evidence level is 1 of 5 (Highest · pooled evidence). Pooled evidence from several studies sits at the top of the evidence pyramid.
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.