已知信号的模糊神经网络识别的硬件实现研究

REALIZATION OF HARDWARE FUZZY NEURAL NETWORK RECOGNITION OF KNOWN SIGNALS

为了提高模糊神经网络识别已知信号的速度 ,本文提出了采用多片TMS32 0C40构成的并行处理系统实现该方法的具体方案 .在分析算法结构的基础上 ,给出了由 5片C40构成的并行处理系统实现并行算法的原理和方法 ,该并行处理系统充分利用了C40的结构特点 ,采用星型结构 ,使处理系统的处理速度更快 ,从而提高了模糊神经网络方法的识别速度 .试验仿真结果表明 ,该系统的加速比和效率合理 ,并行性能良好 .

Presents the multi_TMS320C40 parallel processing system,which makes full use of the structural characteristics of C40 and improves the processing speed by using the starlike structure that the recognition speed of known signal is improved,and concludes from experimental results that this system has rational acceleration ratio and efficiency, and good parallel performance.