Work

Ultra lightweight AES S-box ASIC hardware for low-power IoT devices

Year: 2026

Type: article

Abstract: In pursuit of secure, resource-constrained cryptographic hardware for the Internet of Things (IoT), this work presents an Advanced Encryption Standard (AES) Substitution-box (S-box) architecture optimized for ultra-low power, minimal area, and high throughput. The proposed design leverages composite Galois Field (CGF) arithmetic over $\textrm{GF}(((2^4)^2))$ , incorporating isomorphic and inverse-isomorphic transformations to reduce inversion complexity in $\textrm{GF}(2^8)$ . A unified transformation engine merges affine and inverse-isomorphic operations, significantly minimizing logic redundancy. The architecture is based on purely combinational logic, thereby eliminating the need for sequential control and enabling single-propagation-cycle substitution. Post-layout static timing analysis indicates that the design achieves a maximum operating frequency of 892.86 MHz, which equates to a throughput of 7.14 Gbps. Logic minimization, substructure reuse, and shared multiplier-XOR networks are applied to optimize gate-level complexity and energy consumption further. Physical design implementation and validation were conducted using the Synopsys digital design flow with a $32~\textrm{nm}$ CMOS standard cell library, achieving a compact core area of $411~\mathrm {\mu m}^2$ and total power of just $192~\mathrm {\mu W}$ . The resulting hardware efficiency reaches 17.39 Mbps/ $\mu$ m $^{2}$ , indicating improved area-to-throughput efficiency under comparable implementation conditions. This design demonstrates that advanced algebraic decomposition, linear transformation fusion, and combinational datapath modeling can effectively address the growing demand for lightweight, fabrication-ready cryptographic primitives. The architecture serves as a hardware-efficient implementation suitable for deployment in secure IoT edge devices, smart sensors, and energy-constrained embedded systems.

Source: Scientific Reports

DOI: 10.1038/s41598-026-69487-2

Authors: Salman Ahmed, Nabihah Ahmad, Abedalmuhdi M. AlMomany, Chessda Uttraphan, Uzair Soomro, Abdinasir Hirsi Abdi

Institutions: Tun Hussein Onn University of Malaysia, Gulf University for Science & Technology

Language: English

Topic: Cryptographic Implementations and Security

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: gold

Estimated APC paid: $2,690