The anticipated emergence of a cryptographically-relevant quantum computer (CRQC) would fundamentally compromise much of the security used in public key cryptography today. Traditional cryptographic systems, commonly used for key exchanges and digital signatures, are therefore vulnerable. Efforts to standardize and migrate to post-quantum cryptographic (PQC) alternatives require rigorous empirical evaluation to ensure that these new methods in which we place our security are not vulnerable to quantum computers or less secure than the traditional systems that they aim to replace. This thesis explores one method of combining the security of today’s key exchanges with that of the newly standardized post-quantum alternatives using a hybrid approach within the Transport Layer Security (TLS) 1.3 protocol. This was implemented into MbedTLS, and tests were conducted to evaluate the impact of different hybrid key combinations in a simulated network environment with restrictive Maximum Transmission Unit (MTU) sizes and higher levels of packet loss. The results indicated that the TCP segmentation introduced by the limiting MTU was the greatest factor affecting the TLS 1.3 handshake latency when performed over lossy networks.
The anticipated emergence of a cryptographically-relevant quantum computer (CRQC) would fundamentally compromise much of the security used in public key cryptography today. Traditional cryptographic systems, commonly used for key exchanges and digital signatures, are therefore vulnerable. Efforts to standardize and migrate to post-quantum cryptographic (PQC) alternatives require rigorous empirical evaluation to ensure that these new methods in which we place our security are not vulnerable to quantum computers or less secure than the traditional systems that they aim to replace. This thesis explores one method of combining the security of today’s key exchanges with that of the newly standardized post-quantum alternatives using a hybrid approach within the Transport Layer Security (TLS) 1.3 protocol. This was implemented into MbedTLS, and tests were conducted to evaluate the impact of different hybrid key combinations in a simulated network environment with restrictive Maximum Transmission Unit (MTU) sizes and higher levels of packet loss. The results indicated that the TCP segmentation introduced by the limiting MTU was the greatest factor affecting the TLS 1.3 handshake latency when performed over lossy networks.
The emergence of quantum computing presents a significant threat to classical cryptographic primitives, particularly those employed in securing internet communications via widely used protocols such as Transport Layer Security (TLS). As conventional key exchange mechanisms will become increasingly vulnerable in the post-quantum era, the integration of post-quantum cryptographic (PQC) algorithms into existing security protocols is of utmost importance. This study investigates the impact of incorporating PQC key encapsulation mechanisms—specifically, the recent standards CRYSTALS-Kyber and HQC, in conjunction with the candidate standard BIKE—into the TLS 1.3 handshake. A comprehensive experimental evaluation was conducted to measure handshake latency under emulated network conditions with varying packet loss probabilities. The findings offer useful insights into the performance trade-offs introduced by PQC integration and further highlight the necessity of a timely transition to post-quantum cryptographic standards.
The emergence of large-scale quantum computers poses a significant threat to current cryptographic protocols, particularly Transport Layer Security (TLS), which relies on RSA and elliptic curve cryptography vulnerable to Shor's algorithm. This paper presents a hybrid framework for quantum-resistant TLS handshakes that integrates post-quantum cryptography (PQC), quantum key distribution (QKD) and classical cryptography to provide comprehensive protection against both classical and quantum attacks. The approach employs three QKD protocols; BB84, E91, and Measurement-Device-Independent QKD (MDI-QKD) alongside NIST-standardized PQC algorithms as well as alternative candidates such as HQC and BIKE. A key innovation is the QKD key management module, which addresses performance bottlenecks through periodic key generation and secure reuse within predefined retention periods, significantly reducing QKD overhead in TLS handshakes. The experimental implementation using the quantum network simulator evaluated 144 distinct cryptographic protocol combinations. Results demonstrate that ML-KEM combined with ML-DSA provides optimal PQC performance, while QKD protocol selection is distance-dependent: BB84 for short-range and MDI-QKD for long-distance communication. The framework includes various attack mitigation mechanisms and ensure security as long as at least one cryptographic method remains uncompromised. Consistent with current global quantum network usage, the proposed TLS approach is designed primarily for MAN and WAN environments rather than the Internet, providing a balanced solution that combines theoretical security guarantees with computational efficiency.
This thesis investigates the current state of Post-quantum cryptography (PQC) in the industry, with a particular focus on its practical deployment, integration challenges, and performance implications.The study combines a structured analysis of academic literature, standardization efforts, and industry guidance with an experimental evaluation of PQC integration in widely used security protocols. The analysis shows that, while significant progress has been made in the standardization of quantum-resistant algorithms such as ML-KEM, ML-DSA, and SLH-DSA, real-world adoption remains in an early and transitional phase. Organizations primarily focus on evaluation and pilot implementations, as deployment is constrained by interoperability issues, implementation maturity, and the complexity of migrating existing cryptographic infrastructures.To complement the conceptual analysis, experimental measurements were conducted for TLS, SSH, and IPsec to evaluate the performance impact of hybrid post-quantum cryptographic mechanisms. The results indicate that performance overhead is not uniform across components. In particular, post-quantum signature schemes introduce significant computational and network overhead, mainly due to increased key and certificate sizes. In contrast, hybrid key exchange mechanisms show only moderate performance impact across all evaluated protocols.The findings highlight that the primary challenge of PQC adoption lies not in algorithm design but in practical integration and large-scale migration. Hybrid cryptographic approaches emerge as the most viable strategy for enabling a gradual transition, balancing improved security against quantum threats with operational feasibility.Overall, this thesis demonstrates that PQC is transitioning from a theoretical research field to a practical engineering challenge, with significant implications for the future design of secure communication systems.
Quantum computers could break currently used asymmetric cryptographic schemes in a few years using Shor's algorithm. They are used in numerous protocols and applications to secure authenticity as well as key agreement, and quantum-safe alternatives are urgently needed. NIST therefore initiated a standardization process. This requires intensive evaluation, also with regard to performance and integrability. Here, the integration into TLS 1.3 plays an important role, since it is used for 90% of all Internet connections. In the present work, algorithms for quantum-safe key exchange during TLS 1.3 handshake were reviewed. The focus is on the influence of dedicated network parameters such as transmission rate or packet loss in order to gain insights regarding the suitability of the algorithms under corresponding network conditions. For the implementation, a framework by Paquin et al. was extended to emulate network scenarios and capture the handshake duration for selected algorithms. It is shown that the evaluated candidates Kyber, Saber and NTRU as well as the alternative NTRU Prime have a very good overall performance and partly undercut the handshake duration of the classical ECDH. The choice of a higher security level or hybrid variants does not make a significant difference here. This is not the case with alternatives such as FrodoKEM, SIKE, HQC or BIKE, which have individual disadvantages and whose respective performance varies greatly depending on the security level and hybrid implementation. This is especially true for the data-intensive algorithm FrodoKEM. In general, the prevailing network characteristics should be taken into account when choosing scheme and variant. Further it becomes clear that the performance of the handshake is influenced by external factors such as TCP mechanisms or MTU, which could compensate for possible disadvantages due to PQC if configured appropriately.
The rapid advancement of quantum computing threatens the security foundations of today’s communication protocols. On the web, the traditional approach to securing web traffic has been to couple HTTP with TLS over TCP, but QUIC over UDP has recently emerged as an alternative to reduce latency and improve performance over unreliable networks. The need for an urgent transition to quantum-resistant web protocols demands further examination of both stacks. Consequently, in this work we evaluate their performance when combined with post-quantum primitives. To this end, we devise a practical evaluation framework that integrates the cryptographic primitives into TLS and QUIC implementations. Our analysis focuses on comparing the impact of hybrid and post-quantum primitives on TLS and QUIC under both ideal and realistic network conditions, providing quantitative insights into the performance cost and feasibility of transitioning toward a post-quantum web. Our results indicate that hybrid KEMs incur the highest handshake latency and bandwidth overhead, while pure post-quantum KEMs offer a favorable trade-off between security and performance, with only moderate costs. Moreover, QUIC consistently reduces the performance penalty of post-quantum primitives compared to TLS, especially in lossy network environments.
Quantum Computers (QCs) differ radically from traditional computers and can efficiently solve mathematical problems fundamental to our current cryptographic algorithms. Although existing QCs need to accommodate more qubits to break cryptographic algorithms, the concern of ''Store-Now-Decrypt-Later'' (i.e., adversaries store encrypted data today and decrypt them once powerful QCs become available) highlights the necessity to adopt quantum-safe approaches as soon as possible. In this work, we investigate the performance impact of Post-Quantum Cryptography (PQC) on TLS 1.3. Different signature algorithms and key agreements (as proposed by the National Institute of Standards and Technology (NIST)) are examined through black- and white-box measurements to get precise handshake latencies and computational costs per participating library. We emulated loss, bandwidth, and delay to analyze constrained environments. Our results reveal that HQC and Kyber are on par with our current state-of-the-art, while Dilithium and Falcon are even faster. We observed no performance drawback from using hybrid algorithms; moreover, on higher NIST security levels, PQC outperformed any algorithm in use today. Hence, we conclude that post-quantum TLS is suitable for adoption in today's systems.