TY - JOUR
T1 - Implementation of Efficient XOR Arbiter PUF on FPGA With Enhanced Uniqueness and Security
AU - Anandakumar, N. Nalla
AU - Hashmi, Mohammad S.
AU - Chaudhary, Muhammad Akmal
N1 - Funding Information:
This work was supported by the Collaborative Research Grant (CRP) at Nazarbayev University under Grant 021220CRP0222.
Publisher Copyright:
© 2013 IEEE.
PY - 2022
Y1 - 2022
N2 - Physically unclonable functions (PUFs) are recently utilized as a promising security solution for authentication and identification of internet of things (IoT) devices. In this article, we present an efficient implementation of XOR Arbiter PUF (XOR APUF) on field-programmable gate arrays (FPGAs). In our work, we incorporated concept of discrete programmable delays logic (PDL) configurations, the obfuscated challenges and temporal majority voting (TMV) before the XOR operation concurrently to enhance uniqueness and security. The derived design has been verified on 25 Xilinx Artix-7 FPGAs and the results are promising with uniqueness of 48.69%, uniformity 50.73% and reliability 99.41%, which significantly improves over previous work into XOR APUF designs. In addition, we also investigate modeling attack resistance of the proposed design against various modelling attacks and the security analysis show that the proposed PUF has lower prediction rate (<65%) when compared to the previous XOR APUF designs.
AB - Physically unclonable functions (PUFs) are recently utilized as a promising security solution for authentication and identification of internet of things (IoT) devices. In this article, we present an efficient implementation of XOR Arbiter PUF (XOR APUF) on field-programmable gate arrays (FPGAs). In our work, we incorporated concept of discrete programmable delays logic (PDL) configurations, the obfuscated challenges and temporal majority voting (TMV) before the XOR operation concurrently to enhance uniqueness and security. The derived design has been verified on 25 Xilinx Artix-7 FPGAs and the results are promising with uniqueness of 48.69%, uniformity 50.73% and reliability 99.41%, which significantly improves over previous work into XOR APUF designs. In addition, we also investigate modeling attack resistance of the proposed design against various modelling attacks and the security analysis show that the proposed PUF has lower prediction rate (<65%) when compared to the previous XOR APUF designs.
KW - FPGA
KW - Internet of Things (IoT)
KW - Physical unclonable functions (PUFs)
KW - programmable delays line (PDL)
KW - XOR arbiter PUF (XOR APUF)
UR - http://www.scopus.com/inward/record.url?scp=85144746044&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85144746044&partnerID=8YFLogxK
U2 - 10.1109/ACCESS.2022.3228635
DO - 10.1109/ACCESS.2022.3228635
M3 - Article
AN - SCOPUS:85144746044
SN - 2169-3536
VL - 10
SP - 129832
EP - 129842
JO - IEEE Access
JF - IEEE Access
ER -