Research Journal of Information Technology1815-74322151-7959Science International10.3923/rjit.2009.1.16PonalagusamyR. SenthilkumarS. 1200911In 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.]]>Bader, M.,198711121124Bader, M.,198899215219Butcher, J.C.,1987Pages: 512Pages: 512Chua, L.O. and L. Yang,19883512571272Chua, L.O. and L. Yang,19883512731290Gonzalez, R.C., R.E. Woods and S.L. Eddin,20052005Lai, K.K. and P.H.W. Leong,19951995pp: 5154Lee, C.C. and J.P. De-Gyvez,19946217220Mitsui, T. and Y. Shinohara,1995Nossek, J.A., G. Seiler, T. Roska and L.O. Chua,199220533553Oliveira, S.C.,199920185187Ponalagusamy, R. and S. Senthilkumar,20072007Roska et al.,19941994