Décio Luiz Gazzoni Filho
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Décio Luiz Gazzoni Filho has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Julio López
UNICAMP
3 shared publications - Caio Teixeira
UNICAMP
2 shared publications - Félix Carvalho Rodrigues
UNIVESP / UNICAMP
2 shared publications - Guilherme Brandão Da Silva
Independent Researcher
2 shared publications - Ricardo Dahab
UNICAMP
2 shared publications - Daniel Strufaldi Batista
State University of Londrina - UEL
1 shared publication
Latest Publications
- Open Access
- 8 July 2024
- Published by International Association for Cryptologic Research in IACR Communications in Cryptology
- Vol. 1 (2)
Efficient isochronous fixed-weight sampling with applications to NTRU
We present a solution to the open problem of designing a linear-time, unbiased and timing attack-resistant shuffling algorithm for fixed-weight sampling. Although it can be implemented without timing leakages of secret data in any architecture, we illustrate with ARMv7-M and ARMv8-A implementations; for the latter, we take advantage of architectural features such as NEON and conditional instructions, which are representative of features available on architectures targeting similar systems, such as Intel. Our proposed algorithm improves asymptotically upon the current approach based on constant-time sorting networks ( O ( n ) versus O ( n log 2 n ) ), and an implementation of the new algorithm applied to NTRU is also faster in practice, by a factor of up to 6.91 ( 591 % ) on ARMv8-A cores and 12.89 ( 1189 % ) on the Cortex-M4; it also requires fewer uniform random bits. This translates into performance improvements for NTRU encapsulation, compared to state-of-the-art implementations, of up to 50% on ARMv8-A cores and 72% on the Cortex-M4, and small improvements to key generation (up to 2.7% on ARMv8-A cores and 6.1% on the Cortex-M4), with negligible impact on code size and a slight improvement in RAM usage for the Cortex-M4.
- 8 July 2024
- Published by International Association for Cryptologic Research in IACR Communications in Cryptology
- Vol. 1 (2)
Open AccessEfficient isochronous fixed-weight sampling with applications to NTRU
We present a solution to the open problem of designing a linear-time, unbiased and timing attack-resistant shuffling algorithm for fixed-weight sampling. Although it can be implemented without timing leakages of secret data in any architecture, we illustrate with ARMv7-M and ARMv8-A implementations; for the latter, we take advantage of architectural features such as NEON and conditional instructions, which are representative of features available on architectures targeting similar systems, such as Intel. Our proposed algorithm improves asymptotically upon the current approach based on constant-time sorting networks ( O ( n ) versus O ( n log 2 n ) ), and an implementation of the new algorithm applied to NTRU is also faster in practice, by a factor of up to 6.91 ( 591 % ) on ARMv8-A cores and 12.89 ( 1189 % ) on the Cortex-M4; it also requires fewer uniform random bits. This translates into performance improvements for NTRU encapsulation, compared to state-of-the-art implementations, of up to 50% on ARMv8-A cores and 72% on the Cortex-M4, and small improvements to key generation (up to 2.7% on ARMv8-A cores and 6.1% on the Cortex-M4), with negligible impact on code size and a slight improvement in RAM usage for the Cortex-M4.
- Open Access
- 9 April 2024
- Published by International Association for Cryptologic Research in IACR Communications in Cryptology
- Vol. 1 (1)
Fast polynomial multiplication using matrix multiplication accelerators with applications to NTRU on Apple M1/M3 SoCs
Efficient polynomial multiplication routines are critical to the performance of lattice-based post-quantum cryptography (PQC). As PQC standards only recently started to emerge, CPUs still lack specialized instructions to accelerate such routines. Meanwhile, deep learning has grown immeasurably in importance. Its workloads call for teraflops-level of processing power for linear algebra operations, mainly matrix multiplication. Computer architects have responded by introducing ISA extensions, coprocessors and special-purpose cores to accelerate such operations. In particular, Apple ships an undocumented matrix-multiplication coprocessor, AMX, in hundreds of millions of mobile phones, tablets and personal computers. Our work repurposes AMX to implement polynomial multiplication and applies it to the NTRU cryptosystem, setting new speed records on the Apple M1 and M3 systems-on-chip (SoCs): polynomial multiplication, key generation, encapsulation and decapsulation are sped up by 1.54 – 3.07 × , 1.08 – 1.33 × , 1.11 – 1.50 × and 1.20 – 1.98 × , respectively, over the previous state-of-the-art.
- 9 April 2024
- Published by International Association for Cryptologic Research in IACR Communications in Cryptology
- Vol. 1 (1)
Open AccessFast polynomial multiplication using matrix multiplication accelerators with applications to NTRU on Apple M1/M3 SoCs
Efficient polynomial multiplication routines are critical to the performance of lattice-based post-quantum cryptography (PQC). As PQC standards only recently started to emerge, CPUs still lack specialized instructions to accelerate such routines. Meanwhile, deep learning has grown immeasurably in importance. Its workloads call for teraflops-level of processing power for linear algebra operations, mainly matrix multiplication. Computer architects have responded by introducing ISA extensions, coprocessors and special-purpose cores to accelerate such operations. In particular, Apple ships an undocumented matrix-multiplication coprocessor, AMX, in hundreds of millions of mobile phones, tablets and personal computers. Our work repurposes AMX to implement polynomial multiplication and applies it to the NTRU cryptosystem, setting new speed records on the Apple M1 and M3 systems-on-chip (SoCs): polynomial multiplication, key generation, encapsulation and decapsulation are sped up by 1.54 – 3.07 × , 1.08 – 1.33 × , 1.11 – 1.50 × and 1.20 – 1.98 × , respectively, over the previous state-of-the-art.
- Open Access
- 11 September 2023
- Published by Universidade Estadual de Londrina in Semina: Ciências Exatas e Tecnológicas
- Vol. 44, e48268
Real-Time Ego-Lane Detection in a Low-Cost Embedded Platform using CUDA-Based Implementation
This work assesses the effectiveness of heterogeneous computing based on a CUDA implementation for real-time ego-lane detection using a typical low-cost embedded computer. We propose and evaluate a CUDA-optimized algorithm using a heterogeneous approach based on the extraction of features from an aerial perspective image. The method incorporates well-known algorithms optimized to achieve a very efficient solution with high detection rates and combines techniques to enhance markings and remove noise. The CUDA-based solution is compared to an OpenCV library and to a serial CPU implementation. Practical experiments using TuSimple's image datasets were conducted in an NVIDIA's Jetson Nano embedded computer. The algorithm detects up to 97.9% of the ego lanes with an accuracy of 99.0% in the best-evaluated scenario. Furthermore, the CUDA-optimized method performs at rates greater than 300 fps in the Jetson Nano embedded system, speeding up 25 and 140 times the OpenCV and CPU implementations at the same platform, respectively. These results show that more complex algorithms and solutions can be employed for better detection rates while maintaining real-time requirements in a typical low-power embedded computer using a CUDA implementation.
- 11 September 2023
- Published by Universidade Estadual de Londrina in Semina: Ciências Exatas e Tecnológicas
- Vol. 44, e48268
Open AccessReal-Time Ego-Lane Detection in a Low-Cost Embedded Platform using CUDA-Based Implementation
This work assesses the effectiveness of heterogeneous computing based on a CUDA implementation for real-time ego-lane detection using a typical low-cost embedded computer. We propose and evaluate a CUDA-optimized algorithm using a heterogeneous approach based on the extraction of features from an aerial perspective image. The method incorporates well-known algorithms optimized to achieve a very efficient solution with high detection rates and combines techniques to enhance markings and remove noise. The CUDA-based solution is compared to an OpenCV library and to a serial CPU implementation. Practical experiments using TuSimple's image datasets were conducted in an NVIDIA's Jetson Nano embedded computer. The algorithm detects up to 97.9% of the ego lanes with an accuracy of 99.0% in the best-evaluated scenario. Furthermore, the CUDA-optimized method performs at rates greater than 300 fps in the Jetson Nano embedded system, speeding up 25 and 140 times the OpenCV and CPU implementations at the same platform, respectively. These results show that more complex algorithms and solutions can be employed for better detection rates while maintaining real-time requirements in a typical low-power embedded computer using a CUDA implementation.
- 12 September 2022
- Published by Sociedade Brasileira de Computação
- p. 29-42
Evaluation of modular multiplication techniques for Supersingular Isogeny Schemes on ARMv8 cores
This paper focuses on the evaluation of different modular multiplication implementation techniques on 64-bit ARMv8 systems for the third-round NIST alternate candidate SIKE. The benchmarks were performed on four devices: an Orange Pi WinPlus featuring the Cortex-A53 processor, an NVIDIA Jetson Nano with a Cortex-A57, a Raspberry Pi 4 with a Cortex-A72 and a Macbook Air based on an Apple M1 chip. Throughout these platforms we observed that the two-level Karatsuba Comba method performs better on most Cortex-A processors but the Operand Scanning method presented a performance improvement ranging from 10% to 43% for the multiplication procedure and a 7% to 25% improvement for the modular reduction on the Apple M1 for all SIKE security levels, resulting in an overall improvement ranging from 8% to 28% for the SIKE KEM operations on this architecture.
- 12 September 2022
- Published by Sociedade Brasileira de Computação
- p. 29-42
Evaluation of modular multiplication techniques for Supersingular Isogeny Schemes on ARMv8 cores
This paper focuses on the evaluation of different modular multiplication implementation techniques on 64-bit ARMv8 systems for the third-round NIST alternate candidate SIKE. The benchmarks were performed on four devices: an Orange Pi WinPlus featuring the Cortex-A53 processor, an NVIDIA Jetson Nano with a Cortex-A57, a Raspberry Pi 4 with a Cortex-A72 and a Macbook Air based on an Apple M1 chip. Throughout these platforms we observed that the two-level Karatsuba Comba method performs better on most Cortex-A processors but the Operand Scanning method presented a performance improvement ranging from 10% to 43% for the multiplication procedure and a 7% to 25% improvement for the modular reduction on the Apple M1 for all SIKE security levels, resulting in an overall improvement ranging from 8% to 28% for the SIKE KEM operations on this architecture.
- 12 September 2022
- Published by Sociedade Brasileira de Computação
- p. 372-377
Evaluation of CRYSTALS-Kyber and Saber on the ARMv8 architecture
This paper presents preliminary experimental results for the implementation of the third-round NIST finalists CRYSTALS-Kyber and Saber on the ARMv8 architecture. Our implementation uses NEON instructions to speed up key generation, encapsulation, and decapsulation of the reference codes. The benchmarks are performed on three devices: an Orange Pi WinPlus (Cortex-A53), a Raspberry Pi 4 (Cortex-A72), and a MacBook Air based on an Apple M1 chip. The experimental results show substantial improvements for Kyber and Saber, with speed-ups in the ranges 1.16-1.38x and 1.21-1.96x, respectively. We focused on the most time-consuming operations of each cryptosystem; however, similar works suggest that more expressive speed-ups can be obtained by extending the use of NEON instructions to other primary sub-routines.
- 12 September 2022
- Published by Sociedade Brasileira de Computação
- p. 372-377
Evaluation of CRYSTALS-Kyber and Saber on the ARMv8 architecture
This paper presents preliminary experimental results for the implementation of the third-round NIST finalists CRYSTALS-Kyber and Saber on the ARMv8 architecture. Our implementation uses NEON instructions to speed up key generation, encapsulation, and decapsulation of the reference codes. The benchmarks are performed on three devices: an Orange Pi WinPlus (Cortex-A53), a Raspberry Pi 4 (Cortex-A72), and a MacBook Air based on an Apple M1 chip. The experimental results show substantial improvements for Kyber and Saber, with speed-ups in the ranges 1.16-1.38x and 1.21-1.96x, respectively. We focused on the most time-consuming operations of each cryptosystem; however, similar works suggest that more expressive speed-ups can be obtained by extending the use of NEON instructions to other primary sub-routines.