Fast Fourier transform convolutional neural network accelerator based on overlap addition  

作　　者：You Chen Li Dejian Feng Xi Shen Chongfei Wei Jizeng Liu Yu 

机构地区：[1]School of Microelectronics,Tianjin University,Tianjin 300072,China [2]Beijing Smart-Chip Microelectronics Technology Company Limited,Beijing 102200,China [3]College of Intelligence and Computing,Tianjin University,Tianjin 300072,China

出　　处：《The Journal of China Universities of Posts and Telecommunications》2024年第5期71-84,共14页中国邮电高校学报（英文版）

基　　金：supported by the Project of the State Grid Corporation of China in 2022 (5700-201941501A-0-0-00);the National Natural Science Foundation of China (U21B2031)。

摘　　要：In convolutional neural networks(CNNs), the floating-point computation in the traditional convolutional layer is enormous, and the execution speed of the network is limited by intensive computing, which makes it challenging to meet the real-time response requirements of complex applications. This work is based on the principle that the time domain convolution result equals the frequency domain point multiplication result to reduce the amount of floating-point calculations for convolution. The input feature map and the convolution kernel are converted to the frequency domain by the fast Fourier transform(FFT), and the corresponding point multiplication is performed. Then the frequency domain result is converted back to the time domain, and the output result of the convolution is obtained. In the shared CNN, the input feature map is much larger than the convolution kernel, resulting in many invalid operations. The overlap addition method is proposed to reduce invalid calculations and speed up network execution better. This work designs a hardware accelerator for frequency domain convolution and verifies its efficiency on the Xilinx Zynq UltraScale+MPSoC ZCU102 board. Comparing the calculation time of visual geometry group 16(VGG16) under the ImageNet dataset faster than the traditional time domain convolution, the hardware acceleration of frequency domain convolution is 8.5 times.

关 键 词：convolutional neural network(CNN) fast Fourier transform(FFT) overlap addition 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]