指数函数的fpga实现及算法优化

Fpga Implementation and Algorithm Optimization of Exponential Function

采用切比雪夫多项式逼近指数函数,并用modelsim进行仿真验证。针对“积之和”的函数形式,采用分布式算法对计算过程进行优化,提高运算速度。用流水线方式实现运算过程。通过缩小LUT设计规模,减小芯片面积,采用并行运算进一步提高运算速度。

Exponential function is widely used. Chebyshev polynomial is used to approximate the exponential function,and the simulation is carried out with Modelsim. For the function form of "sum of products", the distributed algorithm is used to optimize the calculation process and improve the operation speed. Pipeline method is used to realize the operation process.By reducing the design scale of LUT, the chip area is reduced. The parallel algorithm is adopted to further improve the operation speed.