A 0.61-μJ/Frame Pipelined Wired-logic DNN Processor in 16-nm FPGA Using Convolutional Non-Linear Neural Network.
Atsutake KosugeYao-Chung HsuMototsugu HamadaTadahiro KurodaPublished in: IEEE Open J. Circuits Syst. (2022)
Keyphrases
- neural network
- parallel architecture
- high speed
- single chip
- systolic array
- network architecture
- gate array
- training process
- logic circuits
- highly non linear
- wireless networks
- digital signal
- hardware implementation
- parallel processing
- modal logic
- back propagation
- low power
- data flow
- frame rate
- fpga device
- artificial neural networks
- xilinx virtex
- field programmable gate array
- distributed memory
- power reduction
- random access memory
- deep learning
- neural network model
- video frames
- signal processing
- fuzzy logic
- real time
- multi core processors
- instruction set
- level parallelism
- chip design
- logic programming
- application specific integrated circuits