Secure Aggregation in Federated Learning for Non-IID Data Environments

Privacy-Preserving Federated Learning Framework for Robust Model Training Under Non IID Data Distributions

Federated learning is a decentralised machine-learning approach in which several clients jointly build a shared model without moving their raw data to one location. Rising concerns around privacy, tightening regulation, and restrictions on how data may be owned or shared have made this approach increasingly attractive in practice. Although federated learning lowers privacy exposure relative to centralised training, deploying it in practice is complicated by clients whose data are unevenly distributed and non-identically distributed, by clients that participate inconsistently, and by training that can converge unpredictably. To obtain global models that train reliably and consistently even when client data are heterogeneous, this work puts forward a federated learning system built around privacy preservation. The design follows a client–server pattern in which a coordinating server aggregates updates from local models using weights that account for imbalance among participants. The behaviour of the resulting system is examined methodically across several data-distribution regimes — IID, mildly non-IID, and severely non-IID. The experiments show that the framework converges reliably and delivers predictive accuracy that holds up well, especially in the more difficult non-IID cases. Compared with conventional federated learning baselines, the approach shows greater robustness and steadier performance across successive training rounds. Because it is simple to implement, repeatable, and built with real deployment in mind, the architecture suits privacy-sensitive, decentralised use cases such as distributed intelligent systems, industrial monitoring, and healthcare analytics.

Decentralized and Secure Aggregation in Federated Learning

Federated learning (FL) is a machine learning technique where multiple clients with local data collaborate in training a machine learning model. However, this centralization of sensitive model updates incurs security and privacy issues. A malicious aggregator can launch inference attacks to violate the privacy of clients' private data from local models and explore sensitive knowledge from global model. Trusted Execution Environment (TEE) based schemes offer secure enclaves to protect models privacy during the aggregation procedure. However, TEE technologies incur new functional and security assumptions, where a single vulnerability could undermine the trusted system. Cryptographic schemes such as homomorphic encryption (HE), differential privacy (DP) and secret sharing (SS) cannot achieve the simultaneous requirements of efficiency, accuracy and privacy currently. In this paper, we present DoshFL that avoids expensive cryptographic operations and enables tunable trade-off between privacy and efficiency in federated learning for model asset protection.

SoK: Verifiable Integrity Claims for Privacy-Preserving Federated Learning

This SoK model federated learning as an append-only transcript of submissions, admissions, aggregation, and finalization events, and formalize verifiability as a collection of integrity claims issued by clients and the aggregator, and checked by different verifier classes.

PRoVeFL: Private Robust and Verifiable Aggregation in Federated Learning

This work proposes PRoVeFL-a novel, modular FL framework that is Privacy-preserving, Byzantine-Robust, and ensures Verifiable aggregation, and improves runtime over the prior works, Prio and ELSA, based on distributed trust with comparable security guarantees, up to 100x and 10x, respectively.

NETFL: Network Efficient Fully Secure Federated Learning at Scale

Federated Learning Architectures For Privacy-Preserving Customer Intelligence Across Telecom Networks

Telecom operators hold some of the richest customer data in any industry, yet the models that would turn it into churn, lifetime-value, fraud, and personalization intelligence are held back by a structural problem: the data cannot be pooled. Customer records sit across regional operators, business units, edge networks, and clouds, each under a different privacy regime such as GDPR, CCPA, or telecom-specific governance, and moving them into one repository raises privacy risk, regulatory exposure, transfer cost, and latency. This paper presents a privacy-by-design federated learning reference architecture for customer intelligence across telecom networks, in which each operator trains locally on its own CRM, billing, service-interaction, and network-telemetry data and shares only encrypted model updates, never raw records. The architecture engineers for the realities that break naive federation in telecom: non-independent, non-identically distributed data across regions, addressed with weighted aggregation, local epochs, adaptive learning rates, and client clustering; and heterogeneous, intermittently connected nodes, addressed with a hybrid-asynchronous coordination scheme, aggregation windows, fault-tolerant checkpointing, and client selection. Secure aggregation, differential privacy, and encrypted transport are bound to explicit GDPR, CCPA, and telecom-governance evidence so data locality is auditable, not merely asserted. We report projected performance ranges and an illustrative simulation rather than production benchmarks, and we state that limit plainly. The contribution is the reference architecture, its privacy-assurance model, a decision framework for when federation earns its complexity in telecom, and an evaluation plan for the empirical phase.