fnt

A Parameterized FPGA Realization of High - Speed FNT Processor

一种可参数化快速FNT的FPGA实现

This paper presents a method of building expert systems with fuzzy-neural technology ' ( FNT method ) .

本文提出了一种用模糊－神经技术建造专家系统的方法（FNT方法）。

Because of FNT 's inherent restrictions , it is hard to implement cyclic convolution using FNT .

由于FNT变换本身的局限，不能完成大点数循环卷积。

For fixed-size formats like FNT or PCF , each image is a bitmap .

对于固定尺寸字体格式，如FNT或者PCF，每一个图像都是一个位图。

A Convolver Implemented by Using FNT

一个用FNT实现的卷积器

Its advantages include not only zero-error brought by FNT , but also convenience for hardware manipulation , no saving for intermediate data and less area-consume .

不仅具有FNT本身零误差的优点，而且有适合于硬件实现，具有不用保存中间结果，节省面积的优点。

On the basis of FPT , in this paper we introduce an iteration algorithm for computing hyper-large scale 2-D cyclic convolutions by combining FPT with FNT .

本文将FPT和FNT相结合提出了一种计算超大型二维循环卷积的迭代算法，它的基础是应用了FPT。

These works are presented in this thesis : 1 , The multiplication-free Fermat number transform ( FNT ) with the cyclic convolution property is used for calculating the multiplication of two polynomials to improve speed and accuracy .

本文有特色的主要工作如下：1、利用Fermat数变换（FNT）的循环卷积性质进行整系数多项式相乘，提高了运算的精度和速度。

In order to find a better network structure , this article first uses a flexible neural tree ( FNT ) in protein tertiary structure prediction , PSO optimizes the network parameters , probabilistic incremental program evolution ( PIPE ) optimizes the network structure .

为了寻找一种更优的网络结构，本文首次将灵活神经树（FNT）应用在蛋白质三级结构预测中，PSO对网络参数进行优化，概率增强式程序进化（PIPE）对网络结构进行优化。

This paper describes the basic concept and circuit structure of FNT , demonstrates the method by which we implemented a 8 point 5 bit FNT convolver with a base 2 pipeline architecture , and introduces a kind of new code fitting for FNT hardware logic .

叙述了FNT的基本概念和电路结构，说明了用FNT实现一个8点5位基2流水线工作方式卷积器的方法，介绍了适用于FNT硬件逻辑的新码。

As compared with the FPT algorithm for 2-D cyclic convolutions , the amount of multiplications of the algorithm introduced here decreases by one order of magnitude , and also it relieves the strict restriction of word length about cyclic convolution scale in FNT algorithm of 2-D cyclic convolutions .

与二维循环卷积的FPT算法相比，乘法量减少了一个数量级，同时实际取消了FNT算法中卷积规模所受到的字长的限制。