1. Title of the Publication Designing Hardware-Friendly Hash Functions for Network Security Using Cartesian Genetic Programming ----------------------------------------------------------------------------------------------------------------------- 2. Author Information Mujtaba Hassan{1} - mujtaba.hassan@kuleuven.be Jo Vliegen{1} - jo.vliegen@kuleuven.be Stjepan Picek{3} - stjepan.picek@ru.nl Nele Mentens{1,2} - nele.mentens@kuleuven.be 1. ES&S, COSIC, ESAT, KU Leuven, Belgium 2. LIACS, Leiden University, The Netherlands 3. Digital Security Group, Radboud University, The Netherlands ----------------------------------------------------------------------------------------------------------------------- Corresponding Author Name: Mujtaba Hassan email: mujtaba.hassan@kuleuven.be ----------------------------------------------------------------------------------------------------------------------- 4. Paper Abstract In this study, we propose a novel, hardware-efficient non-cryptographic (NC) hash function, developed using Cartesian Genetic Programming (CGP), to optimize the processing of network flows (e.g., 96-bit vectors in IPv4). With the advent of high-speed terabit Ethernet technologies such as 800G, cybercriminals are increasingly exploiting these networks to launch various attacks, including distributed denial-of-service (DDoS) attacks. To counter these threats, network security applications often employ probabilistic data structures (PDS), such as Bloom filters and Count Min sketches, to monitor network flows. These applications are often deployed on Field Programmable Gate Arrays (FPGAs) to meet real-time processing demands. The performance of PDS depends significantly on the efficiency of NC-hash functions. By utilizing avalanche metrics (avalanche dependence, avalanche weight, and entropy) as the fitness function, CGP-hash ensures robust performance without the need for dataset-specific training. While Genetic Programming (GP)-based NC-hash functions have demonstrated superior computational efficiency on FPGAs in terms of operating frequency, throughput, and latency, they are often less resource-efficient compared to state-of-the-art NC-hash functions. Thus, our hypothesis in this paper is that CGP, with its compact representation, leads to NC hashes with high computational efficiency and fewer resources on an FPGA. Our experimental results confirm that CGP-hash improves computational efficiency by at least 7.3% while improving area efficiency by 4.5x compared to state-of-the-art NC-hash functions. This makes CGP-hash more suitable for FPGA implementations than other bio-inspired and handcrafted hash functions. Moreover, the 48-bit hash output can be extended by evolving additional hash functions and concatenating their outputs, all while maintaining superior resource efficiency and computational speed. ----------------------------------------------------------------------------------------------------------------------- 5. A list containing one or more of the eight letters (A, B, C, D, E, F, G, or H) that correspond to the criteria that the author claims that the work satisfies (B) The result is equal to or better than a result that was accepted as a new scientific result at the time when it was published in a peer-reviewed scientific journal. (D) The result is publishable in its own right as a new scientific result independent of the fact that the result was mechanically created. (E) The result is equal to or better than the most recent human-created solution to a long-standing problem for which there has been a succession of increasingly better human-created solutions. (F) The result is equal to or better than a result that was considered an achievement in its field at the time it was first discovered. ----------------------------------------------------------------------------------------------------------------------- 6. Statement Why the Results Satisfy the Criteria (B) The result is equal to or better than a result that was accepted as a new scientific result at the time when it was published in a peer-reviewed scientific journal. Ans: Our methodology surpasses all known FPGA implementations of state-of-the-art Non Cryptographic (NC) hash functions, which rely on avalanche metrics. In comparison against 18 other NC hash functions, CGP-hash achieves a computational efficiency improvement of at least 7.3% in terms of operating frequency, throughput, and latency while improving area efficiency (Throughput/Look up Tables (LUT)) by 4.5 times. This efficiency enables our approach to scale to larger hash outputs through result concatenation, supporting network applications that require extended hash sizes without significant FPGA resource consumption. (D) The result is publishable in its own right as a new scientific result independent of the fact that the result was mechanically created. Ans: The scientific merit of CGP-hash, making it a publishable result independent of its evolutionary origin, is founded on the following key advancements it offers to the field of network security applications and hash function design: 1: Demonstrate superior performance and resource usage on FPGA 2: Inherently compact and hardware-friendly structure 3: Robustness and practical scalability for Probabilistic data structures These demonstrated characteristics, specifically the novel achievement of concurrent high computational speed and drastically improved hardware resource usage on FPGA, packaged in a compact and robust design, constitute a significant and practical advancement in the design of NC hash functions for critical, high-speed network security applications. Such a result, offering tangible benefits and clear advantages over existing solutions, would be considered a publishable scientific contribution in its own right within reputable peer-reviewed journals or conferences focused on network engineering, FPGA/hardware design, or computer security. The value lies in the superior hash function itself and its direct applicability, independent of the fact that Cartesian Genetic Programming was the technique employed for its discovery. (E) The result is equal to or better than the most recent human-created solution to a long-standing problem for which there has been a succession of increasingly better human-created solutions. Ans: Our result improves upon the most recent human-designed solutions to the long-standing and actively evolving challenge of optimizing non-cryptographic (NC) hash functions for use in Probabilistic Data Structures (PDS) on Terabit Ethernet networks (bandwidth ≥100 Gb/s). Previous approaches, including reduced-round cryptographic functions (e.g., XOODOO-NC, SPECK-NC, GIFT-NC, PHOTON-NC), CRC-based methods, and other manually crafted NC hash functions (e.g., Murmur3, FNV1-a, SipHash, XORHash), have steadily advanced in performance over time. Our CGP-evolved hash function outperforms all of them in latency, throughput, and area efficiency, setting a new state-of-the-art and continuing the clear trend of iterative improvement in this field. (F) The result is equal to or better than a result that was considered an achievement in its field at the time it was first discovered. The CGP-hash function exceeds previous results that were widely regarded as milestones in the field such as FNV1-a, Murmur3, SipHash, XORHash and XOODOO-NC in terms of throughput, latency, operating frequency and resource usage on FPGA. In particular, it also improves upon evolutionary designs of NC hash functions such as GPNCH, NCGPH96-concat and E-hash, which demonstrated leading results for their FPGA performance in terms of computational efficiency. By delivering higher operating frequency, lower latency, and significantly better area efficiency, CGP-hash establishes a new performance benchmark. These measurable improvements over previously celebrated results reinforce its qualification under this criterion. ----------------------------------------------------------------------------------------------------------------------- 7. Full Citation Hassan, M., Vliegen, J., Picek, S., Mentens, N. (2025). Designing Hardware-Friendly Hash Functions for Network Security Using Cartesian Genetic Programming. In: García-Sánchez, P., Hart, E., Thomson, S.L. (eds) Applications of Evolutionary Computation. EvoApplications 2025. Lecture Notes in Computer Science, vol 15612. Springer, Cham. https://doi.org/10.1007/978-3-031-90062-4_14 ----------------------------------------------------------------------------------------------------------------------- 8. Prize Money Breakdown We share the money equally. ----------------------------------------------------------------------------------------------------------------------- 9. A statement stating why the authors expect that their entry would be the "best," We believe our entry is the best because it presents a scientifically novel, highly efficient, and practically impactful solution to a real-world challenge of growing urgency: detecting malicious network traffic at line rate on Terabit Ethernet networks (bandwidth ≥100 Gb/s). A key bottleneck in such systems is ensuring that network monitoring and filtering keep pace with line speed, a failure to do so results in packet loss and security blind spots. Our CGP-evolved hash function delivers unprecedented area efficiency on FPGA, achieving a throughput-per-LUT (Tp/LUT) of 450.43 Mbps/LUT, over 4.5 times better than leading alternatives like XOODOO-NC. It also surpasses prior evolutionary and non-evolutionary methods in terms of computational speed (operating frequency, throughput, and latency) on FPGA. Beyond performance, our solution offers scalability and generalizability. The CGP-hash structure can be extended by concatinating the outputs, to generate larger outputs without compromising efficiency, making it well-suited for modern probabilistic data structures like Bloom filters used in security-critical applications. This combination of scientific innovation, measurable technical excellence, and practical deployment potential in cloud, telecom, and cybersecurity contexts makes our entry to stand out from both academic and practical perspectives. ----------------------------------------------------------------------------------------------------------------------- 10. Evolutionary Computation Type Genetic Programming (Cartesian Genetic Programming) ----------------------------------------------------------------------------------------------------------------------- 11. Publication Date 17 April 2025