2026-05-22T16:05:55Zhttps://keep.lib.asu.edu/oai/request

oai:keep.lib.asu.edu:node-1982362024-12-23T18:01:48Zoai_pmh:alloai_pmh:repo_items

198236 https://hdl.handle.net/2286/R.2.N.198236 http://rightsstatements.org/vocab/InC/1.0/ All Rights Reserved 2024 137 pages Masters Thesis Academic theses Text eng Ahmed, Anees Kinsy, Michel Ehret, Alan Trieu, Ni Arizona State University Partial requirement for: M.S., Arizona State University, 2024 Field of study: Computer Science Zero Knowledge Proofs (ZKP) represent a major development in the field of cryptography and privacy-preserving computation. ZKP-powered digital infrastructure demands minimal trust from users while prioritizing their data privacy. Despite the existence of numerous open-source tools and a growing need for privacy protection in various industries, the integration of ZKP protocols into practical platforms has been notably scarce. Two primary challenges faced by developers are proficiency in the complex mathematical concepts that underpin these protocols, and the high computational costs associated with their implementation. For my thesis research, I developed a novel acceleration framework, named ZKXL (Zero-Knowledge "X-celeration" Library), with the aim of reducing the computational complexity and facilitating the integration of Zero Knowledge Proof (ZKP) protocols into practical, real-world applications. ZKXL is a library of composable hardware primitives for accelerating ZKP computations and improving their computational efficiency, with a particular focus on cost-effective low-power FPGA (field-programmable gate array) platforms. As part of my evaluation efforts, I conducted a comprehensive analysis of 23 open-source software libraries that implement ZKP protocols. This involves assessing their respective advantages and limitations, conducting benchmark tests on their operational efficiency, and identifying potential avenues for further investigation and improvement. Furthermore, I created a set of illustrative code snippets for all the components in the ZKP library, serving as a useful reference tool for developers exploring ZKP technology. In experimental evaluations using MiMC hashing, ZKXL shows an 11x increase in throughput while consuming 79x less power than server-grade CPUs. ZKXL is accessible as an open-source solution, easily deployable, and serves as a valuable tool for organizations and researchers engaged in zero-knowledge proof applications. The primary contributions of this thesis work are to lower the development difficulty and enable the efficient deployment of zero-knowledge proof-based applications by both industry developers and researchers. Computer Science FPGA Hardware Acceleration MiMC Software Survey Zero Knowledge Proof Hardware Primitives Library for Acceleration of Zero-Knowledge Proof Algorithms and Applications