Implementation of 4-way Superscalar Hash MIPS Processor Using FPGA

• Published in: Journal of physics. Conference series Vol. 1003; no. 1; pp. 12037 - 12044
• Main Authors: Omran, Safaa Sahib, Jumma, Laith Fouad
• Format: Journal Article
• Language: English
• Published: Bristol IOP Publishing 01.05.2018
• Subjects: Algorithms; BLAKE; Cryptography; Field programmable gate arrays; FPGA; Hardware description languages; Hash based algorithms; Integrated circuits; Microprocessors; MIPS; Personal communications systems; Physics; Security; SHA-1; SHA-2; VHSIC (circuits); Wireless communications
• ISSN: 1742-6588; 1742-6596
• DOI: 10.1088/1742-6596/1003/1/012037

Due to the quick advancements in the personal communications systems and wireless communications, giving data security has turned into a more essential subject. This security idea turns into a more confounded subject when next-generation system requirements and constant calculation speed are considered in real-time. Hash functions are among the most essential cryptographic primitives and utilized as a part of the many fields of signature authentication and communication integrity. These functions are utilized to acquire a settled size unique fingerprint or hash value of an arbitrary length of message. In this paper, Secure Hash Algorithms (SHA) of types SHA-1, SHA-2 (SHA-224, SHA-256) and SHA-3 (BLAKE) are implemented on Field-Programmable Gate Array (FPGA) in a processor structure. The design is described and implemented using a hardware description language, namely VHSIC "Very High Speed Integrated Circuit" Hardware Description Language (VHDL). Since the logical operation of the hash types of (SHA-1, SHA-224, SHA-256 and SHA-3) are 32-bits, so a Superscalar Hash Microprocessor without Interlocked Pipelines (MIPS) processor are designed with only few instructions that were required in invoking the desired Hash algorithms, when the four types of hash algorithms executed sequentially using the designed processor, the total time required equal to approximately 342 us, with a throughput of 4.8 Mbps while the required to execute the same four hash algorithms using the designed four-way superscalar is reduced to 237 us with improved the throughput to 5.1 Mbps.