面向神经形态感知的人工脉冲神经元的研究进展

近年来,随着人工智能技术和脉冲神经网络(SNN)的迅猛发展,人工脉冲神经元的研究逐渐兴起。人工脉冲神经元的研究对于开发具有人类智能水平的机器人、实现自主学习和自适应控制等领域具有重要的应用前景。传统的电子器件由于缺乏神经元的非线性特性,需要复杂的电路结构和大量的器件才能模拟简单的生物神经元功能,同时功耗也较高。因此,最近研究者们借鉴生物神经元的工作机制,提出了多种基于忆阻器等新型器件的人工脉冲神经元方案。这些方案具有功耗低、结构简单、制备工艺成熟等优点,并且在模拟生物神经元的多种功能等方面取得了显著进展。文章将从人工脉冲神经元的基本原理出发,综述和分析目前已有的各种实现方案。具体来说,将分别介绍基于传统电子器件和基于新型器件的人工脉冲神经元的实现方案,并对其优缺点进行比较。此外,还将介绍不同类型的人工脉冲神经元在实现触觉、视觉、嗅觉、味觉、听觉和温度等神经形态感知方面的应用,并对未来的发展进行展望。希望能够为人工脉冲神经元的研究和应用提供有益的参考和启示。     

Feedforward associative memory switched-capacitor artificial neural networks

Observed spatiotemporal firings of biological neurons have lead many researchers to believe that the rate of firings of these biological neurons is what conveys neuronal information in the brain. In this paper we seek to highlight parallels between biological neurons and observed effects in real neurons, with artificial neurons implemented as switched-capacitor structures. One such effect is the heavy use of lateral inhibition observed in the brain that is often modeled by winner-take-all analog circuits. This paper introduces a novel winner-take-all circuit using switched capacitors that truly mimics this effect seen in biological systems. In addition, we show how switched-capacitor structures can also cater to both binary and bipolar coding of input data vectors, as required by many artificial neural network paradigms today. Applications of switched-capacitors artificial neural networks to pattern recognition and character recognition problems using feedforward associative neural networks are also discussed, and two examples are provided. Simulations using both HSPICE and SWITCAP2 confirm all our expectations.     

Analog circuits for modeling biological neural networks: design andapplications

Computational neuroscience is emerging as a new approach in biological neural networks studies. In an attempt to contribute to this field, we present here a modeling work based on the implementation of biological neurons using specific analog integrated circuits. We first describe the mathematical basis of such models, then present analog emulations of different neurons. Each model is compared to its biological real counterpart as well as its numerical computation. Finally, we demonstrate the possible use of these analog models to interact dynamically with real cells through artificial synapses within hybrid networks. This method is currently used to explore neural networks dynamics.     

基于UML自动测试系统软件面向对象建模

分析了自动测试系统的现状,提出运用UML工具,对自动测试系统进行面向对象建模的思想.介绍了对航电模拟器系统软件进行面向对象建模的过程.采用UML中的用例图、类图和序列图,从功能、静态模型和动态模型,3方面对航电模拟器系统软件平台进行了描述.     

Modeling a Mobile Cellular Network Using a Discrete-Event Simulator

In this paper we introduce a discrete-event-based simulation technology suitable to model cellular mobile radio systems with respect to their dynamic behavior. The latter results in the need for simulation tools which support mutable system configurations. The problems encountered when modeling mutable system configurations using conventional discrete-event simulators are discussed. In order to show a possible solution to this problem we introduce configuration events and configuration objects into the theory of discrete-event simulations. We demonstrate how configuration objects can be implemented in a discrete-event simulator using a dynamic map-function, hereby extending a definition of higher order functions. We apply the configuration objects in a Multiple Layer Model for modeling an entire mobile cellular radio network in a discrete-event simulator as an application for the extended theory. In this model we use the configuration objects to change the network's configuration during runtime. We show that this solution—combined with an object-oriented software design and, possibly, a visual programming language—is a powerful tool for the simulation of the dynamic aspects of mobile cellular radio networks. The software technology presented will be applied by the German cellular network operator Mannesmann Mobilfunk to model dynamic features applicable to the radio resource management of mobile radio networks and to access their performance by simulation.     

Synthetic organisms and self-designing systems

This paper examines the need for complex, adaptive solutions to certain types of complex problems typified by the Strategic Defense System and NASA's Space Station and Mars Rover. Since natural systems have evolved with capabilities of intelligent behavior in complex, dynamic situations, it is proposed that biological principles be identified and abstracted for application to certain problems now facing industry, defense, and space exploration. Two classes of artificial neural networks are presented — a nonadaptive network used as a genetically determined “retina,” and a frequency-coded network used as an adaptive “brain.” The role of a specific environment coupled with a system of artificial neural networks having simulated sensors and effectors is seen as an ecosystem. Evolution of synthetic organisms within this ecosystem provides a powerful optimization methodology for creating intelligent systems able to function successfully in any desired environment. A complex software system involving a simulation of an environment and a program designed to cope with that environment are presented. Reliance on adaptive systems, as found in nature, is only part of the proposed answer, though an essential one. The second part of the proposed method makes use of an additional biological metaphor—that of natural selection—to solve the dynamic optimization problems every intelligent system eventually faces. A third area of concern in developing an adaptive, intelligent system is that of real-time computing. It is recognized that many of the problems now being explored in this area have their parallels in biological organisms, and many of the performance issues facing artificial neural networks may find resolution in the methodology of real-time computing.     

Use of hybrid models for testing and debugging control software for electromechanical systems

This paper proposes a hybrid modeling language and its application to a simulator-based testing and debugging environment for the control software for electromechanical systems. The new hybrid modeling language is designed mainly focusing on simulation speed, flexibility in connecting with control software, and model reusability. This language maintains the advantages of existing hybrid modeling languages such as Hybrid cc, including the flexibility of constraint programming and the reusability of the object-oriented approach. A new feature of the language is that it allows combination of compositional constraint programming and sequential procedural programming. The compiled code is executed efficiently by the runtime system, which has a built-in mechanism for communicating with external software, eliminating the complicated setup required for integrating the simulator with the control software. Model components programmed by the object-oriented approach allow designers to use existing components and to concentrate on the task of modeling the newly designed hardware. The runtime system has been integrated with a three-dimensional kinematics simulator and a control software design tool to create a simulator-based testing and debugging environment. The effectiveness of this system has been confirmed through its application to real product design projects.     

Some Insights Into Computational Models of (Patho)physiological Brain Activity

The amount of experimental data concerning physiology and anatomy of the nervous system is growing very fast, challenging our capacity to make comprehensive syntheses of the plethora of data available. Computer models of neuronal networks provide useful tools to construct such syntheses. They can be used to interpret experimental data, generate experimentally testable predictions, and formulate new hypotheses regarding the function of the neural systems. Models can also act as a bridge between different levels of neuronal organization. The ultimate aim of computational neuroscience is to provide a link between behavior and underlying neural mechanisms. Depending on the specific aim of the model, there are different levels of neuronal organization at which the model can be set. Models are constructed at the microscopic (molecular and cellular), macroscopic level (local populations or systems), or dynamical systems level. Apart from purely computational models, hybrid networks are being developed in which biological neurons are connected in vitro to computer simulated neurons. Also, neuromorphic systems are recently being created using silicon chips that mimic computational operations in the brain. This paper reviews various computational models of the brain and insights obtained through their simulations.     

INVESTIGATION ON THE COMPUTATIONAL PROPERTIES OF INTRANEURONAL DYNAMICS

This paper aims at exploring computational properties of dynamic processes in neu-ral systems,studying their mathematical formulation,and applying the results to artificial neuralnetwork modeling.The stimulus-response processes in neurons are first introduced briefly,thenproperties of neurons described by the Hodgkin-Huxley equations are analyzed.After studyinghow to simplify,the Hodgkin-Huxley equations while maintaining its properties,the concept of dy-namic neuron model is proposed.It is pointed out that the neuron model should include internalstates in order to obtain time-variant thresholds,such as refractory periods of neurons.Finallywe discuss problems related to neural network models based on pulse-stream communication andthe contribution of intraneuronal dynamics to collective properties of the neural network.     

FPGA realization of a neural-network-based nonlinear channel equalizer

The software simulation as well as the hardware implementation of equalizers for transmissions through nonlinear communication channels based on artificial neural networks structure is presented in this paper. We consider four-quadrature-amplitude-modulation technique as an example and compare the performance of two different structures of equalizer, namely, the linear least-mean-square-based equalizer (LIN) and the functional link artificial neural networks (FLANN). The learning curve and symbol error rate for the two structures are respectively evaluated by computer simulation. Besides, the systems have been implemented using field-programmable-gate-array devices. As FLANN uses functions to expand the dimensionality of the input signals, it has about the same system complexity as LIN. But FLANN can achieve fast processing speed under parallel processing structure. Simulation results have demonstrated that FLANN presents much better error performance than LIN, especially when the communication channel is highly nonlinear.     

Redox Memristors with Volatile Threshold Switching Behavior for Neuromorphic Computing

The spiking neural network(SNN), closely inspired by the human brain, is one of the most powerful platforms to enable highly efficient, low cost, and robust neuromorphic computations in hardware using traditional or emerging electron devices within an integrated system. In the hardware implementation, the building of artificial spiking neurons is fundamental for constructing the whole system. However, with the slowing down of Moore’s Law,the traditional complementary metal-oxide-semiconductor(CM...     

单神经元电路的混沌保密通信

人工神经元是构成人工神经网络的基本单元,研究了改进的神经元权值电路的构成,提出了基于模拟电子技术的人工神经元电路的混沌保密通信系统,应用EWB进行了电路实验仿真,设计了保密通信模拟电路系统,实验结果表明设计方案克服了混沌掩盖系统混沌信号能量小和占用频带宽等弱点,提高了通信系统的保密性能。     

A concurrent object-oriented switching program in Chill

A switching program structure that is expandable to allow inclusion of new service features and functions with minimum effort during run time, adaptable to handle enhanced services, allows new services to be added quickly, and is highly reliable, is proposed. It is based on the use of object-oriented programs, hierarchically structured programs, and building-block switching systems based on distributed processing. The discussion covers the logical switching system model, object-oriented switching program systems, concurrent object-oriented programming in Chill (CCITT high level languages), the software architecture, the building-block switching systems, and call processing. Early evaluation results are included     

Role Models — Patterns of Agent System Analysis and Design

Due to their autonomy and social behaviour, agents will play important roles in emerging enterprises; they will fill key positions and provide essential capabilities. This paper describes role modelling as a software engineering technique for specifying, analysing, and designing systems on the basis of the roles that the agents will play. This paper builds on our earlier research in patterns of agent systems. A pattern is a useful solution to a reoccurring problem in a particular context; patterns can be used to solve software engineering problems in analysis, design, and implementation. This paper explains how object-oriented role models can be extended to represent patterns of agent interaction that can then be employed to engineer agent systems. Role theory deals with collaboration and co-ordination. Roles have also been applied to distributed systems management, and to agent and robot systems. However, there has to date not been a methodology for realising these representations in an automated or semi-automated system, due to the lack of adequate formalism and corresponding abstractions in software. Role models rectify this situation for software analysis, design, and implementation.     

Modelling of LDMOS transistor using artificial neural networks: DC and RF performances

This article presents an accurate method based on artificial neural networks (ANNs) for DC and RF modelling of laterally diffused metal oxide semiconductor (LDMOS) transistors, under various temperature conditions. In LDMOS transistors, temperature is an effective factor, so the proposed models include this parameter. Two neural networks‐based procedures have been proposed for LDMOS transistor modelling, first for DC and second for RF modelling. In each case, two kinds of neural networks have been used, multilayer perceptron and radial basis function neural networks. Two models are compared to each other in terms of accuracy, and for both of them, an excellent agreement between modelled and measured data is obtained. The ANN model is developed and trained with the help of data obtained by simulation of a Si‐LDMOS transistor using ADS software.     

Spike Based Normalizing Hebbian Learning in an Analog VLSI Artificial Neuron

Recent neurophysiological results indicate that changes in synaptic efficacy are dependent on co-occurrence of a pre and a postsynaptic spike at the synapse [5,8]. There are only a few models of parts of the nervous system that use temporal correlation of single spikes in learning [1]. In most models of artificial neural networks neurons communicate by analog signals representing frequencies, and their learning rules are also defined on these continuous signals. Timing of single spikes is not used, nor is it represented. This simplification has proven useful in many applications and it makes simulations in software simpler and faster. Spiking systems have been avoided because they are computationally more difficult. However, by implementing spiking and learning artificial neurons in analog VLSI it is possible to examine the behavior of these more detailed models in real time. This is why ourselves and others [4] have started to use silicon models of spiking learning neurons. We have formulated one possible mechanism of weight normalization: a Hebbian learning rule that makes use of temporal correlations between single spikes. We have implemented it on a CMOS chip and demonstrate its normalizing behavior.     

基于计算智能的信息处理技术的研究与应用

计算智能是人工智能的重要分支,以数据为基础,主要借鉴连接主义和行为主义的思想,基于生物进化和细胞网络等机制,具有分布、并行、自适应、自组织和自学习等特点.首先介绍了计算智能的起源和概念,然后以人工神经网络、遗传算法、蚁群算法为例阐述了其原理和应用,最后介绍了在新技术条件下,计算智能的发展趋势及有待解决的一些问题.     

Recent Advances in Transistor‐Based Artificial Synapses

Simulating biological synapses with electronic devices is a re‐emerging field of research. It is widely recognized as the first step in hardware building brain‐like computers and artificial intelligent systems. Thus far, different types of electronic devices have been proposed to mimic synaptic functions. Among them, transistor‐based artificial synapses have the advantages of good stability, relatively controllable testing parameters, clear operation mechanism, and can be constructed from a variety of materials. In addition, they can perform concurrent learning, in which synaptic weight update can be performed without interrupting the signal transmission process. Synergistic control of one device can also be implemented in a transistor‐based artificial synapse, which opens up the possibility of developing robust neuron networks with significantly fewer neural elements. These unique features of transistor‐based artificial synapses make them more suitable for emulating synaptic functions than other types of devices. However, the development of transistor‐based artificial synapses is still in its very early stages. Herein, this article presents a review of recent advances in transistor‐based artificial synapses in order to give a guideline for future implementation of synaptic functions with transistors. The main challenges and research directions of transistor‐based artificial synapses are also presented.     

一种新型雷达视频模拟器的设计与实现

对雷达视频模拟器的类型、结构、功能等作了系统的分析，重点讨论了一种新型雷达视频模拟器的设计与实现。该模拟器采用“PC插卡”的结构形式，采用ISA总线、内存直接映像和FPGA等技术，具有极大的灵活性和通用性。只需要改变计算机的仿真软件，就可以实现对不同体制雷达信号的模拟。该模拟器经过适当扩展后，可应用于雷达信号环境一体化仿真系统。