An Ultra-High-Throughput MultiCore QDI Asynchronous Hardware Architecture for AES Encryption

This paper presents a multi-core asynchronous AES-128 encryption architecture designed for ultra-high-throughput applications. The proposed design uses quasi-delay-insensitive (QDI) dual-rail logic and a four-phase return-to-zero handshake protocol, eliminating the need for a global clock. Several AES cores share a common pipelined key-expansion unit to reduce the overall area and power overhead. Within each core, throughput is improved using a deeply pipelined 31-stage architecture. The architecture is described at the register-transfer level using Verilog and synthesized in a 15 nm CMOS technology. The results show that a single core can achieve a throughput of up to 107 Gbit/s, while ten cores operating in parallel can provide a throughput of up to 1.07 Tbit/s. The proposed architecture is therefore well suited for high-speed communication standards such as IEEE 802.3bs-2017, which supports data rates of up to 400 Gbit/s.

Authors

Institutions

Publication Details

Journal
Chips
Published
2026-11-13
DOI
https://doi.org/10.3390/chips5040040
Primary Topic
Cryptographic Implementations and Security
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

An Ultra-High-Throughput MultiCore QDI Asynchronous Hardware Architecture for AES Encryption

Ghislain Fraidy Bouesse, Franck Moukanda Mbango
Chips
Cryptographic Implementations and Security
article

An Ultra-High-Throughput MultiCore QDI Asynchronous Hardware Architecture for AES Encryption

Ghislain Fraidy Bouesse, Franck Moukanda Mbango
article en

Abstract

This paper presents a multi-core asynchronous AES-128 encryption architecture designed for ultra-high-throughput applications. The proposed design uses quasi-delay-insensitive (QDI) dual-rail logic and a four-phase return-to-zero handshake protocol, eliminating the need for a global clock. Several AES cores share a common pipelined key-expansion unit to reduce the overall area and power overhead. Within each core, throughput is improved using a deeply pipelined 31-stage architecture. The architecture is described at the register-transfer level using Verilog and synthesized in a 15 nm CMOS technology. The results show that a single core can achieve a throughput of up to 107 Gbit/s, while ten cores operating in parallel can provide a throughput of up to 1.07 Tbit/s. The proposed architecture is therefore well suited for high-speed communication standards such as IEEE 802.3bs-2017, which supports data rates of up to 400 Gbit/s.

ChipsVol. 5(4)
Marien Ngouabi University (CG)
Industry, innovation and infrastructure
Cryptographic Implementations and Security
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon