Federated Learning (FL) enables collaborative machine learning without centralizing raw data. However, the impending threat of cryptographically-relevant quantum computers exposes FL's reliance on classical public-key cryptography (e.g., RSA, ECC) to Harvest-Now-Decrypt-Later (HNDL) attacks. We present PQ-FL, the first open federated learning control plane secured end-to-end using the finalized NIST Post-Quantum Cryptography (PQC) standards. Our architecture integrates ML-DSA-87 (FIPS 204) for transport-layer identity authentication (mTLS) and ML-KEM-1024 (FIPS 203) for application-layer model payload encryption. To overcome the lack of comprehensive PQC support in standard Python libraries (BoringSSL), we introduce a novel OQS-stunnel sidecar pattern that proxies edge client traffic. We evaluate the system on a CUDA-accelerated edge cluster training a convolutional neural network. Empirical results confirm successful multi-round aggregation using FedAvg, Krum, and Trimmed-Mean strategies alongside differential privacy, achieving seamless PQC key encapsulation (1568-byte ciphertexts) and verifiable asynchronous global model streaming.
Federated Learning (FL) enables collaborative machine learning without centralizing raw data. However, the impending threat of cryptographically-relevant quantum computers exposes FL's reliance on classical public-key cryptography (e.g., RSA, ECC) to Harvest-Now-Decrypt-Later (HNDL) attacks. We present PQ-FL, the first open federated learning control plane secured end-to-end using the finalized NIST Post-Quantum Cryptography (PQC) standards. Our architecture integrates ML-DSA-87 (FIPS 204) for transport-layer identity authentication (mTLS) and ML-KEM-1024 (FIPS 203) for application-layer model payload encryption. To overcome the lack of comprehensive PQC support in standard Python libraries (BoringSSL), we introduce a novel OQS-stunnel sidecar pattern that proxies edge client traffic. We evaluate the system on a CUDA-accelerated edge cluster training a convolutional neural network. Empirical results confirm successful multi-round aggregation using FedAvg, Krum, and Trimmed-Mean strategies alongside differential privacy, achieving seamless PQC key encapsulation (1568-byte ciphertexts) and verifiable asynchronous global model streaming.
Industrial innovations, driven by technologies such as the Internet of Things (IoT), cloud computing, and Artificial Intelligence (AI), have revolutionized various operational domains, including supply chain management, equipment monitoring, predictive maintenance, and quality control. Deep Learning (DL) has become instrumental in addressing complex, data-driven challenges within industrial settings, relying on large datasets collected from heterogeneous sources. However, centralizing these datasets on cloud platforms poses significant privacy and security risks. Federated Learning (FL) offers a promising solution by enabling distributed model training across multiple nodes while preventing the exchange of raw data. Despite its advantages, FL introduces new vulnerabilities, especially related to the security of communication channels between participating entities. Traditional cryptographic mechanisms, such as digital signatures and model encryption, can mitigate these risks, but the emergence of quantum computing threatens the robustness of conventional solutions. This work explores the integration of Post-Quantum Cryptography (PQC) into FL to enhance security without incurring significant performance degradation. A modular FL architecture with three security levels is proposed: (i) an unprotected baseline; (ii) authenticated communication using ML-DSA digital signatures; and (iii) full protection combining ML-DSA and ML-KEM-512-based key encapsulation. Implemented within the Flower framework, the architecture is evaluated under simulated MitM attacks. Experimental results demonstrate that PQC-enhanced schemes effectively mitigate quantum-resistant threats while maintaining acceptable computational overhead, thereby ensuring model integrity and data confidentiality.
Collaborative threat intelligence via federated learning (FL) faces critical risks from quantum computing, which can compromise classical encryption methods. This study proposes a quantum-secure FL framework using post-quantum cryptography (PQC) to protect cross-organizational data sharing. We expose vulnerabilities in traditional FL through simulated quantum attacks on RSA encrypted gradients and introduce a hybrid architecture integrating NIST-standardized algorithms CRYSTALS-Kyber for key exchange and CRYSTALS-Dilithium for authentication. Testing on APT attack datasets demonstrated 97.6% threat detection accuracy with minimal latency overhead (18.7%), validating real-world viability. A healthcare consortium case study confirmed secure ransomware indicator sharing without breaching privacy regulations. The work highlights the urgency of quantum ready defenses and provides technical guidelines for deploying PQC in FL systems, alongside policy recommendations for standardizing quantum resilience in threat-sharing networks.
Federated Learning (FL) enables collaborative machine learning by aggregating local models from decentralized clients without sharing raw data. However, the parameter exchange process in standard FL architectures is vulnerable to emerging quantum computing threats. Specifically, Shor's algorithm, executable on large-scale quantum computers, can break conventional asymmetric cryptography such as RSA and ECDH in polynomial time, thereby threatening the security of FL systems that rely on these traditional public-key infrastructures for parameter exchange. This study addresses this vulnerability by implementing the recently standardized post-quantum cryptographic protocol, ML-KEM (Module-Lattice-Based Key-Encapsulation Mechanism), within a FedAvg-based FL architecture. The integration is designed to secure the parameter exchange pipeline without compromising the neural network's performance. Experimental results on a simulated environment utilizing the MNIST dataset demonstrate that the ML-KEM integration preserves 100% of the global model accuracy. Furthermore, the cryptographic latency overhead introduced by encapsulation and decapsulation remains highly efficient, proving its potential as a robust security layer. While current evaluations focus on small-scale deployments, the proposed architecture establishes a foundational post-quantum security framework for future privacy-preserving distributed learning systems.
Federated learning (FL) has become one of the standard approaches for deploying machine learning models on edge devices, where private training data are distributed across clients, and a shared model is learned by aggregating locally computed updates from each client. While this paradigm enhances communication efficiency by only requiring updates at the end of each training epoch, the transmitted model updates remain vulnerable to malicious tampering, posing risks to the integrity of the global model. Although current digital signature algorithms can protect these communicated model updates, they fail to ensure quantum security in the era of large-scale quantum computing. Fortunately, various post-quantum cryptography algorithms have been developed to address this vulnerability, especially the three NIST-standardized algorithms - Dilithium, FALCON, and SPHINCS+.In this work, we empirically investigate the impact of these three NIST-standardized PQC algorithms for digital signatures within the FL procedure, covering a wide range of models, tasks, and FL settings. Our results indicate that Dilithium stands out as the most efficient PQC algorithm for digital signature in federated learning. Additionally, we offer an in-depth discussion of the implications of our findings and potential directions for future research.
The convergence of quantum computing and the scale of Internet of Things (IoT) deployments creates a compounded security problem: classical public-key cryptography securing today’s IoT traffic is vulnerable to harvest-now, decrypt-later attacks, while the constrained hardware typical of IoT devices limits the security mechanisms can be deployed. Prior work has largely evaluated post-quantum cryptography (PQC), federated learning (FL)-based anomaly detection, and quantum-resilient storage separately or in partial combinations, leaving open whether these mechanisms remain practical when composed on constrained hardware under a shared operational budget. This paper answers that question with a three-layer architecture — a hybrid PQC pipeline (Kyber-based key encapsulation, ML-DSA signatures, AES-GCM encryption, and PQC-TLS transport), a federated anomaly-detection layer, and a quantum-resilient storage and audit layer (DeSSE and MuReQua) — executed end-to-end on a Raspberry Pi 5 edge node under AWS IoT Greengrass Core, with nine concurrent virtual IoT nodes processing real attack traffic from the N-BaIoT and TON_IoT datasets. Benchmarking across 27 cryptographic algorithm combinations (810 pipeline executions) shows that all 27 configurations completed in under 2.1 ms, with the deployed Kyber768/ML-DSA-65/AES-256-GCM tier adding 4,521 bytes of per-message overhead, which remained compatible with the evaluated 20 ms inter-row transmission budget. Live federated deployment across the nine virtual IoT nodes on N-BaIoT achieves F1-scores above 0.998 on eight of nine devices, processing at approximately 900 rows/second with zero signature verification failures; a centralized baseline evaluation further confirms 99.46% attack recall on unseen TON_IoT threats withheld from training. The quantum-resilient storage layer maintains 100% chain integrity and complete signature coverage across 4,471 audit records, with independent tamper simulations reliably detected. PQC-TLS introduces a one-time handshake overhead of 13.7 ms (288.5% relative increase) over classical TLS, which remains small relative to the multi-minute FL round intervals measured in the evaluated deployment. Together, these results demonstrate that quantum-resilient cryptography, federated intelligence, and auditable storage can be jointly executed on constrained IoT edge hardware as a single experimentally demonstrated pipeline within the operational conditions evaluated in this study.
As federated learning (FL) keeps dominating the way in privacy-preserving machine learning (ML), notably in finance, its long-term sustainability is confronted with the approaching challenge of quantum computing. The introduction of quantum computing poses a challenge to traditional cryptographic technique on which FL relies RSA, elliptic curve cryptography (ECC), and some homomorphic encryption (HE) methods applied for encrypting model updates during training rendering this chapter to explore the interplay between the threat of quantum computing and the FL architectures, with emphasis on the implications this interaction introduces to cybersecurity and privacy in finance applications. Quantum computers are theoretically capable of solving hard math problems that have been regarded as computationally intractable for quantum computers. The well-known algorithms like Shor’s and Grover’s provides serious risk to the cryptographic procedures FL pipelines depends on. The finance sector, by its nature involving sensitive data and dependence on secure, distributed ML systems, is particularly vulnerable. The chapter begins by summarizing the fundamentals of FL and current encryption methods adapted to protect FL workflows. It then shifts towards the essence of quantum computing, emphasizing on the notable algorithms that threaten FL. We also explore potential attack points through which quantum computing could attack FL, e.g., model inversion attacks, poisoning attacks facilitated by quantum capabilities, and threats to secure aggregation. Post-quantum cryptography (PQC) is also presented as a potential counterattack, followed by a discussion of its use case in FL platforms. This chapter attempts to close the gap between FL and quantum computing threats to the financial sector. It is an introduction to practitioners and researchers to revisit examining security assumptions in FL and move towards the quantum-augmented cyber threat model.