A Hardware Implementation of SOM Neural Network Algorithm

摘要

Aiming at the real-time problem of SOM(Self-Organizing Feature Map) neural network algorithm based on FPGA(Field Programmable Gate Array), this paper analyzes the parallelism and resource reuse of algorithm process, presents a parallel architecture to implement the algorithm, and gives the hardware design method of the key modules in the architecture. It is simulated by Modelsim and runs on the FPGA development board. The experimental results show that the architecture can ensure the performance of neural network and the circuit has a high processing speed.