Open Access Open Access  Restricted Access Subscription Access

Efficient VHDL Simulation for Convolutional Neural Nets via Parallel Pipelining

Niharika Kumari

Abstract


Artificial Neural Networks are computer devices that mimic the way the human brain processes information to handle various tasks. Neural networks are currently widely employed to solve issues in many different disciplines, such as advanced mechanics, photo processing, design acknowledgment, and so on. Given the deep learning computations, there are a lot of new, speedy advances in applications. A deep learning algorithm that extends Fake Brain Organizations and is widely used for image categorization and evidence differentiation. The field for equipment support implies the ever-increasing quantity of handling required by CNNs. Additionally, the streaming nature of CNN operations makes them ideal for reprogrammable hardware architectures like FPGAs. As of yet, no Brain Organization-based approach exists for matching pipes, which is thought to support the existence of a deep brain network in terms of accuracy. By using equal pipelining, the Figuring-based convolutional brain network structure in this paper provided a very precise and practical framework. Developments of the convolutional and fully related layers are understood and considered at the neuronal level. Assessed figuring streamlining implies are observed fractionally at the organizational level by maintaining the organization's precision. In order to optimize the time delay and power consumption of the system, the proposed convolutional brain network is compared to previous regular brain network executions that were carried out on an FPGA using a figuring technique. The results show that the suggested network performs more accurately than the previous ones.


Full Text:

PDF

References


Hamdan, M. K. (2018). VHDL auto-generation tool for optimized hardware acceleration of convolutional neural networks on FPGA (VGT) (Doctoral dissertation, Iowa State University).

Hamdan, M. K., & Rover, D. T. (2017, December). VHDL generator for a high performance convolutional neural network FPGA-based accelerator. In 2017 International Conference on ReConFigurable Computing and FPGAs (ReConFig) (pp. 1-6). IEEE.

Solis, R. E., & Shashev, D. V. (2022). DEVELOPMENTS ON NEURAL NETWORK IMPLEMENTATION FOR FPGA. Интеллектуальные системы 4-й промышленной революции: сборник материалов IV Международного форума, г. Томск, 15–16 декабря 2021 г./под ред. ВИ Сырямкина, 73.

Reggiani, E., Del Sozzo, E., Conficconi, D., Natale, G., Moroni, C., & Santambrogio, M. D. (2021). Enhancing the scalability of multi-fpga stencil computations via highly optimized hdl components. ACM Transactions on Reconfigurable Technology and Systems (TRETS), 14(3), 1-33.

Gschwend, D. (2020). Zynqnet: An fpga-accelerated embedded convolutional neural network. arXiv preprint arXiv:2005.06892.

Samudre, P. H. (2019). Efficient VHDL Simulation of Convolutional Neural Network Using Parallel Pipelining. Journal of VLSI Design and its Advancement, 2(2).

Garland, J., & Gregg, D. (2020). HOBFLOPS CNNs: Hardware Optimized Bitslice-Parallel Floating-Point Operations for Convolutional Neural Networks. arXiv preprint arXiv:2007.06563.

Li, S., Luo, Y., Sun, K., Yadav, N., & Choi, K. K. (2020). A novel FPGA accelerator design for real-time and ultra-low power deep convolutional neural networks compared with titan X GPU. IEEE Access, 8, 105455-105471.

Rout, S. K., Sahani, M., Dora, C., Biswal, P. K., & Biswal, B. (2022). An efficient epileptic seizure classification system using empirical wavelet transform and multi-fuse reduced deep convolutional neural network with digital implementation. Biomedical Signal Processing and Control, 72, 103281.

Bjerge, K., Schougaard, J., & Larsen, D. (2020). A generic and efficient convolutional neural network accelerator using HLS for a system on chip design. CoRR.


Refbacks

  • There are currently no refbacks.