
09/04/2025
๐ Featured Archived Article โ IEEE OJCAS (Vol. 3)
A 0.61-ฮผJ/Frame Pipelined Wired-Logic DNN Processor in 16-nm FPGA Using Convolutional Non-Linear Neural Network
This article presents a wired-logic deep neural network (DNN) processor implemented in a 16-nm FPGA, achieving ultra-low energy consumption and high throughput.
Key contributions include:
๐น A Convolutional Non-Linear Neural Network (CNNN) enabling scalable training and object recognition
๐น A pipeline-type neuron cell that reduces hardware requirements with compact LUTs
๐น Energy efficiency improvements 100ร greater than state-of-the-art binary neural network accelerators
๐ Read the full article in the IEEE OJCAS archive: https://loom.ly/glBn9xA