EUVIMEDEUVIMED entdecken
Crossref

FPGA-DLAF: An FPGA-Based Deep Learning Acceleration Framework for High-Performance and Energy-Efficient Intelligent Computing

Kondoju Pravalika Kondoju Pravalika, Thota Niharika Thota Niharika, Prof A K Rahtod Prof A K Rahtod

International Journal of Science, Strategic Management and Technology · 10.07.2026 · 02 (07) · 1-8

DOI: 10.55041/ijsmt.v2i7.024 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Science, Strategic Management and Technology, online erschienen am 10.07.2026. Verfasst von Kondoju Pravalika Kondoju Pravalika und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The rapid advancement of deep learning has revolutionized numerous application domains including computer vision, natural language processing, healthcare analytics, autonomous systems, and intelligent edge computing. Despite their remarkable performance, deep learning models require extensive computational resources and memory bandwidth, making their deployment challenging in power-constrained and real-time environments. Traditional processing platforms such as Central Processing Units (CPUs) often fail to meet the performance requirements of modern neural networks, while Graphics Processing Units (GPUs) may consume substantial amounts of power and incur high operational costs. Field Programmable Gate Arrays (FPGAs) have emerged as a promising alternative due to their reconfigurability, parallel processing capabilities, and superior energy efficiency. This paper presents an FPGA-based acceleration framework for deep learning models that integrates parallel computation engines, optimized memory architectures, dynamic resource allocation mechanisms, and low-latency inference pipelines. The proposed architecture is designed to accelerate convolutional neural networks and other deep learning models while minimizing hardware resource utilization and power consumption. Experimental analysis demonstrates significant improvements in inference speed, throughput, and energy efficiency compared to conventional computing platforms. The proposed FPGA acceleration framework offers a scalable and flexible solution for deploying deep learning applications in cloud environments, edge devices, and embedded intelligent systems.

Zitieren

APA

Pravalika, K. P. K., Niharika, T. N. T., & Rahtod, P. A. K. R. P. A. K. (2026). FPGA-DLAF: An FPGA-Based Deep Learning Acceleration Framework for High-Performance and Energy-Efficient Intelligent Computing. International Journal of Science, Strategic Management and Technology, 02(07), 1-8. https://doi.org/10.55041/ijsmt.v2i7.024

Vancouver

Pravalika KPK, Niharika TNT, Rahtod PAKRPAK. FPGA-DLAF: An FPGA-Based Deep Learning Acceleration Framework for High-Performance and Energy-Efficient Intelligent Computing. International Journal of Science, Strategic Management and Technology. 2026;02(07):1-8. doi:10.55041/ijsmt.v2i7.024

BibTeX
@article{Pravalika2026,
  author = {Pravalika, Kondoju Pravalika Kondoju and Niharika, Thota Niharika Thota and Rahtod, Prof A K Rahtod Prof A K},
  title = {FPGA-DLAF: An FPGA-Based Deep Learning Acceleration Framework for High-Performance and Energy-Efficient Intelligent Computing},
  journal = {International Journal of Science, Strategic Management and Technology},
  year = {2026},
  volume = {02},
  number = {07},
  pages = {1--8},
  doi = {10.55041/ijsmt.v2i7.024}
}
Kostenlos starten →