摘要：In this research, a versatile algorithm for simulating CNN arrays and time multiplexing is implemented using numerical integration algorithms. The approach, time-multiplexing simulation, plays a pivotal role in the area of simulating hardware models and testing hardware implementations of CNN. Owing to hardware limitations in practical sense, it is not possible to have a one-one mapping between the CNN hardware processors and all the pixels of the image. The simulator provides a solution by processing the input image block by block, with the number of pixels in a block being the same as the number of CNN processors in the hardware. Simulation results and comparison have also been presented to show the efficiency of the Numerical Integration Algorithms. In this research, RK-eight stage seventh order limiting formulas are implemented and it is found that the RK (7,8) algorithm outperforms well in comparison with the Explicit Euler, RK-Gill, RK-fifth order and RK-sixth order algorithms. A more quantitative analysis has been carried out to clearly visualize the goodness and robustness of the numerical algorithms.
