We proposed a Compton scattering imaging model based on energy spectrum, and introduced a coupled gradient neural network implementation of this imaging problem.
提出了基于能量谱的康普顿散射成像模型，引入一种耦合梯度神经网络实现该模型下的康普顿散射成像。
This paper proposed a neural network implementation method for the direction estimation of signal sources based on spatial partition after describing the basic principle.
本文在介绍了空间分割信号源方向估计方法后，提出了一种空间分割信号源方向估计器的神经网络实现方法。
We proposed a Compton scattering imaging model based on energy spectrum, and introduced a coupled gradient neural network implementation of this imaging problem. Successful results were obtained.
提出了基于能量谱的康普顿散射成像模型，引入一种耦合梯度神经网络实现该模型下的康普顿散射成像。得到了成功的结果。
The storage of neural network weight is an important problem of VLSI implementation to be solved.
神经网络的权值存储是集成电路实现技术中涵待解决的重要问题。
This paper analyzes 2 critical problems in FPGA-based implementation of SOM neural network algorithm: parallelizability and finite word-length effect.
分析了SOM神经网络算法在FPGA实现过程中要考虑的2个主要问题：并行性和有限字长效应。
The physical implementation of neural network follows delays of feedback signal and the application of network requires the study of the stability.
神经网络的硬件实现常伴随反馈信号的延时，网络的应用需要对具有时滞反馈网络稳定性的充分研究。
Neural network possesses many of advantages : simple structure, easy implementation in hardware, the basic parallel computational architecture, etc.
人工神经网络的特点是：结构简单、能够大规模并行、容易用硬件实现等。
Neural network models and an optical implementation of the Perception algorithm for pattern dichotomy are described.
本文阐述了几种神经网络模型和模式两分法中感知算法的一种光学执行过程。
The principle, significance and implementation of instrument intelligence are discussed, which combines the Virtual Instrument (VI) and the Fuzzy Neural Network (FNN) technology.
讨论了利用模糊神经网络技术与虚拟仪器技术相结合实现仪器智能化的原理、意义和方案。
According to a learning algorithm of self organizing neural network for mapping character, a CMOS implementation of its synaptic weight by circuit is presented in this paper.
本文根据自组织特征映射神经网络学习算法，提出了其权值的CMOS实现电路。
Based on the threshold decomposition of a k-level signal, an algorithm of decomposition and reconstruction suitable for implementation of feedforward neural network is proposed.
利用信号的阈值分解，导出适合神经网络实现的信号分解和信号恢复算法。
The communication overhead in parallel implementation of artificial neural network on a multiprocessor system is analyzed in this paper.
文中讨论了在多处理器系统上，用大规模并行处理技术实现人工神经网络时，处理器之间的通信开销问题。
A fuzzy neural network is presented. It is essentially a network implementation of fuzzy logic.
提出的一种模糊神经元网络是模糊逻辑的一种网络结构的实现。
The simulation result shows that it can function correctly with fast recognition speed; therefore, it is suitable for VLSI implementation of neural network.
模拟结果表明其功能正确，具有较高的识别速度，适于神经网络的VLSI实现。
Zhang H. Design of electronic nose based on hardware implementation of BP neural network [d]. Chengdu: Southwest Jiaotong University, 2015.
张航。基于硬件实现BP神经网络的电子鼻设计[d]。成都：西南交通大学，2015。
The method can be extended to more types of neural networks, provide a reliable basis for the neural network hardware implementation .
该方法可以推广至更多类型的神经网络，为神经网络的硬件实现提供了可靠的基础。
CMAC (Cerebellar Model Articulation Controller) is a kind of local learning feed - forward neural network with simple architecture, quick learning convergence and effective implementation.
小脑模型清晰度控制器(CMAC)是一种局部学习前馈网络，结构简单，收敛速度快，易于实现。
Optical implementation methods of one dimensional local interconnection neural network (LINN) for associative memory are proposed, and three optoelectronic system are discussed in this paper.
本文提出了一维局域互联关联存贮的光学实现方法，讨论了可用来实现局域互联网的三种光电混合系统。
In this paper, a design of an all-optical multi-value neural network system for implementation of associative memory is presented.
本文提出一种用于联想识别的全光型多值神经网络系统。
Design and implementation of artificial neural networks based on evolutionary computation can lead to significantly better network performance.
基于进化算法可有效解决神经网络设计和实现中存在的一些问题，使网络具有更优的性能。
Design and implementation of artificial neural networks based on evolutionary computation can lead to significantly better network performance.
基于进化算法可有效解决神经网络设计和实现中存在的一些问题，使网络具有更优的性能。
应用推荐