Discussion on the methodology of neural network hardware design and implementation

As a suitable calculating method for parallel computation, neural networks have been widely used in applications. It has been pointed out that the main problem of neural network hardware implementation lies in three aspects. To solve these problems, the way of hardware implementation of neural networks has been discussed. A novel basic mathematical model of the neuron has been introduced. Using this new model of the neuron, the flexibility of the implemented neural network hardware will be greatly improved.