海豚等动物神经系统处理多普勒信号的一种可能性方案

海豚、蝙蝠等动物具有功能强大的生物声纳，生物神经系统处理多普勒效应的网络结构可能会有多种形式.对这类生物的神经系统信号处理机制及结构的研究，将能为信息处理提供新的视角.根据序空间神经编码理论，构建了一个生物神经回路处理多普勒信号.经计算机仿真可知，该生物神经网络运算速度快、分辨率高，并能处理任意的连续变频率多普勒信号.所研究的仅由1—2个神经元组成的网络结构是最简单的，由于网络只有一级神经元的延时，保证了处理的快速性，在神经元输出10个脉冲的情况下分辨率可达7^-10. 关键词： 多普勒效应 海豚 神经编码 神经脉冲序列     

双耳幅值差确定声源方向的神经信息处理机理研究

神经信息系统实质上是定量系统, 应引起足够重视. 关于神经系统的定量研究方面的报道比较少见. 这一问题将会影响进一步的研究, 如双耳声音定向. 双耳定向是定量测量, 用定性分析的方法无法满足要求. 已有的生理实验发现声音输入信号强度与听觉神经的输出频率存在单调递增关系, 所以本文中声音强度的变化被简化成神经脉冲频率的变化. 本文基于圆映射和符号动力学原理, 建立了神经回路定量模型, 模型中对同侧输入回路采用兴奋性耦合, 对侧输入回路采用抑制性耦合, 并考虑神经元间突触连接的量子释放特征, 采用化学耦合模型实现连接, 用耦合系数表示神经元间的耦合程度. 采用Hodgkin-Huxley模型仿真研究听觉神经回路的输入/输出脉冲序列关系. 在已经仿真过的参数范围, 模型在输入信号变化与输出脉冲频率变化间存在单调递增/递减的关系. 对于单输入单输出的神经元, 采用符号动力学方法进行符号化; 对于多输入单输出的神经元, 采用分析各输出脉冲的产生时间, 判断其变化位置, 从神经脉冲序列中得到对应的两耳声音幅值差变化, 以此定位声源. 随着输出脉冲数的增加, 符号序列的长度增加, 符号序列对输入信号变化敏感, 能够得到较高的测量精度. 仿真结果表明这个模型是定量的, 神经脉冲序列能够区分信号的大小.     

声纳方程的瞬态形式讨论

本文提出用瞬态信号的能谱级代替稳态信号的声强级来定义声纳参数，并以短柱声散射为例，运用数值模拟方法，验证用能谱级定义声纳参数，声纳方程是比较有效的，这为应用声纳方程处理瞬态声信号从而获得目标强度提供了一种有效可行的方法。     

耳朵形变赋予蝙蝠超声波束快速适应的物理机制

蝙蝠具备十分复杂的生物声纳系统,可通过发射和接收超声波识别物体。根据超声波频带变化通常可把蝙蝠分为两类:恒频率蝙蝠和调频率蝙蝠。对于恒频率蝙蝠,波束模式非常简单,是由一个较大幅度主瓣和非常少的小幅度旁瓣组成,因此对探测目标的方向特性敏感度不高;然而对于调频率蝙蝠,波束主要是由一系列的     

DC-DC变换器的符号时间序列描述及模块熵分析

本文提出了一种采用符号时间序列和熵理论分析DC-DC变换器非线性行为的方法.该方法首先用离散时间序列描述非线性连续系统,然后将其转换为由简单字符构成的符号序列,再用信息学方法计算出该符号序列的模块熵,从而得到一种新的可量化的非线性动力学行为统计指标.文中以一阶电压反馈DCM和二阶电流反馈CCM Boost变换器为例进行研究.研究结果表明,模块熵这种粗粒化的统计分析方法,能够量化DC-DC变换器的倍周期分岔和混沌行为,且能够准确地确定混沌行为的发生,是一种尚未在DC-DC变换器中提出的简单、实用的分析方法. 关键词： 符号时间序列 符号动力学 模块熵 Lyapunov指数     

非线性物理破解神经编码机制

信息在神经系统中以脉冲点序列的形式传输和处理，神经系统如何通过脉冲点序列对所表达的信息进行编码，一直是一个谜．人们曾普遍认为，神经可能通过发放率（单位时间发放脉冲出现的次数）对信息编码，也有人猜测，神经可通过脉冲点序列的时序编码．由于神经发放的随机性，这就使得任何编码机制都面临着被表达信息的确定性与表达该信息的信号的随机性的矛盾．通过神经非线性随机动力学模型，文章作者发现，神经点序列的发放率对点序列的时序信息传输的影响，揭示了神经点序列时序信息在神经非线性传输中的随机共振特征．由此预期，并进一步通过听觉心理物理实验观察到，在一定条件下噪声对听觉的增强作用．从而通过非线性物理揭示了听觉时序编码机制的存在以及在时序编码中随机性噪声的积极作用．     

5.1通路环绕声的耳机虚拟重放

提出了一种改进的5.1通路环绕声的耳机虚拟重放信号处理方法。在对现有的信号处理方法进行分析,指出它会产生不自然主观听觉效果后,通过理论和心理声学实验证明,采用HRTF信号处理、环绕声信号去相关等方法,在克服普通耳机声重放的“头中定位”的缺点、虚拟出多环绕扬声器效果的同时,并未带来新的不自然听觉效果。因此信号处理方法可增加听觉上的包围感,从而改善耳机重放的主观听觉效果。     

声纳信号宽容性检测指数的初步分析

自上世纪六十年代以来,信号的宽容性(Robust,Robustness)处理就开始在统计数学与信号处理领域受到关注。由于声纳使用的水声环境的特殊复杂性,从事声纳信号处理和声纳设计的学者对环境适应性处理和宽容性水声信号处理开展了大量的研究。主要的目的是探索水声环境效应以及对声纳信号处理的影响,企图寻求一种适应环境的宽容性信号处理方法。本文初步探讨水声信号处理领域宽容性检测的基本概念,给出一种度量宽容性性能的数量指标。同时推导在不同声速剖面下声纳检测宽容性指标和所使用频率的关系。对文中所提出的理论方法与国内外实际海试的某些实例进行了分析比较。     

应用于感存算一体化系统的多模调控忆阻器

交互式人工智能系统的构建依赖于高性能人工感知系统和处理系统的开发.传统的感知处理系统传感器、存储器和处理器在空间上是分离的,感知数据信息的频繁传输和数据格式转换造成了系统的长延时与高能耗.受生物感知神经系统的启发,耦合感知、存储、计算功能的感存算一体化技术为未来感知处理领域提供了可靠的技术方案.具有感知光、压力、化学物质等能力的忆阻器是应用于感存算一体系统的理想器件.本文从器件层面综述了应用于感存算一体化系统忆阻器的研究方向和研究进展,包括视觉、触觉、嗅觉、听觉和多感官耦合类别,并在器件、工艺与集成、电路系统架构和算法方面指出现阶段的挑战与展望,为未来神经形态感存算一体化系统的发展提供可行的研究方向.     

《人工听觉——新视野》

正听觉让我们可以进行言语交流、音乐欣赏、感受环境等声音感知活动。在正常的听觉通路中,声音的机械振动主要通过耳廓、耳道、鼓膜和听骨链的传导进入耳蜗,振动信号在耳蜗中被转换为神经电信号,电信号被听觉神经系统逐渐上传至中枢,经处理形成听觉。当听觉通路受到破坏时,听觉功能就会产生损伤。大多数听力受损者的听觉通路是在机械振动传导环节出了问题,他们可以通过佩戴助听器来获得声学放大,进而得到听力补偿。然而,对于很多重度以上感     

Studies of phase return map and symbolic dynamics in a periodically driven Hodgkin–Huxley neuron

How neuronal spike trains encode external information is a hot topic in neurodynamics studies.In this paper,we investigate the dynamical states of the Hodgkin–Huxley neuron under periodic forcing.Depending on the parameters of the stimulus,the neuron exhibits periodic,quasiperiodic and chaotic spike trains.In order to analyze these spike trains quantitatively,we use the phase return map to describe the dynamical behavior on a one-dimensional(1D)map.According to the monotonicity or discontinuous point of the 1D map,the spike trains are transformed into symbolic sequences by implementing a coarse-grained algorithm—symbolic dynamics.Based on the ordering rules of symbolic dynamics,the parameters of the external stimulus can be measured in high resolution with finite length symbolic sequences.A reasonable explanation for why the nervous system can discriminate or cognize the small change of the external signals in a short time is also presented.     

Why pinnipeds don't echolocate

Odontocete cetaceans have evolved a highly advanced system of active biosonar. It has been hypothesized that other groups of marine animals, such as the pinnipeds, possess analogous sound production, reception, and processing mechanisms that allow for underwater orientation using active echolocation. Despite sporadic investigation over the past 30 years, the accumulated evidence in favor of the pinniped echolocation hypothesis is unconvincing. We argue that an advanced echolocation system is unlikely to have evolved in pinnipeds primarily because of constraints imposed by the obligate amphibious functioning of the pinniped auditory system. As a result of these constraints, pinnipeds have not developed highly acute, aquatic, high frequency sound production or reception systems required for underwater echolocation. Instead, it appears that pinnipeds have evolved enhanced visual, tactile, and passive listening skills. The evolutionary refinement of alternative sensory systems allows pinnipeds to effectively forage, navigate, and avoid predators under water despite the lack of active biosonar capabilities.     

基于符号动力学的认知事件相关电位的复杂度分析

引入了符号动力学方法分析认知事件相关电位(ERP)的复杂度.以混合模型生成的随机时间序列为例,对近似熵和符号熵作了比较.应用符号熵分析了Oddball范式中不同任务条件(靶刺激和非靶刺激)下的ERP的复杂度.研究发现,额区、中央区和顶区的ERP复杂度在刺激呈现后的任务加工时间段内显著减小(非靶刺激和靶刺激分别在刺激呈现后200—300和400—500ms),而且靶刺激ERP复杂度大约在P300成分的峰值时刻达到最小值,在响应之后逐渐回升.这表明基于符号动力学的复杂度分析能够反映认知任务加工的时间过程,并且 关键词： 事件相关电位 符号动力学 熵     

基于改进的符号转移熵的心脑电信号耦合研究

试图探究动力系统中的耦合关系一直以来都是国内外众多学者关注的热点,传统的时间序列符号化分析方法会使研究结果受序列非平稳性的严重影响,本文在原有转移熵的研究基础上,应用粗粒化提取,经过理论与实验的分析,发现心脑电信号耦合研究中的转移熵值在不同提取情况下对应不同的分布趋势,并选择效果最好的信号数据提取方法用在其后的应用分析中. 此外,对时间序列符号化方法提出改进,采用动态的自适应分割方法. 实验结果表明,无论清醒期还是睡眠期,改进的符号转移熵算法观测分析到的心脑电信号耦合作用更显著,能更好的捕捉到信号中的动态信息、系统动力学复杂性的改变,更利于医学临床实践应用中的检测,在分析非平稳的时间序列上具有更好的效果. 关键词： 心脑电信号 粗粒化 符号转移熵 基本尺度     

Noseleaf furrows in a horseshoe bat act as resonance cavities shaping the biosonar beam

Horseshoe bats emit their ultrasonic biosonar pulses through nostrils surrounded by intricately shaped protuberances (noseleaves). While these noseleaves have been hypothesized to affect the sonar beam, their physical function has never been analyzed. Using numerical methods, we show that conspicuous furrows in the noseleaf act as resonance cavities shaping the sonar beam. This demonstrates that (a) animals can use resonances in external, half-open cavities to direct sound emissions, (b) structural detail in the faces of bats can have acoustic effects even if it is not adjacent to the emission sites, and (c) specializations in the biosonar system of horseshoe bats allow for differential processing of subbands of the pulse in the acoustic domain.     

Source-to-sensation level ratio of transmitted biosonar pulses in an echolocating false killer whale

Transmitted biosonar pulses, and the brain auditory evoked potentials (AEPs) associated with those pulses, were synchronously recorded in a false killer whale Pseudorca crassidens trained to accept suction-cup EEG electrodes and to detect targets by echolocation. AEP amplitude was investigated as a function of the transmitted biosonar pulse source level. For that, a few thousand of the individual AEP records were sorted according to the spontaneously varied amplitude of synchronously recorded biosonar pulses. In each of the sorting bins (in 5-dB steps) AEP records were averaged to extract AEP from noise; AEP amplitude was plotted as a function of the biosonar pulse source level. For comparison, AEPs were recorded to external (in free field) sound pulses of a waveform and spectrum similar to those of the biosonar pulses; amplitude of these AEPs was plotted as a function of sound pressure level. A comparison of these two functions has shown that, depending on the presence or absence of a target, the sensitivity of the whale's hearing to its own transmitted biosonar pulses was 30 to 45 dB lower than might be expected in a free acoustic field.     

Ultrasound beam shift induced by short-beaked common dolphin's(Delphinus delphis) tissues as an attenuating gradient material

To understand sound propagation and beam formation, the physical properties of soft tissues from the biosonar system of odontocetes should be explored. Based on the acoustic impedance distributions of biosonar systems, these processes have been examined via numerical simulations. In this study, the images of a short-beaked common dolphin(Delphinus delphis) were obtained via computed tomography. Then, the dolphin was dissected to extract tissue samples for additional examination. In addition to the speed of sound and density measurements, the acoustic attenuation coefficients of the biosonar system in the forehead were tested. The results revealed that the inner layer of the forehead was characterized using low sound speed, low density, and high attenuation. Acoustic fields and beam patterns were then evaluated by setting acoustic attenuation coefficients at different levels. Sounds propagating along the low-attenuation path had a lesser reduction in amplitude. Beam directivities in near and far fields suggested that changes in attenuation distribution would cause beam patterns to shift. These results indicated the complexity of a dolphin's sonar emission system and helped improve our understanding of sound energy attenuation via studies on the forehead of odontocetes.     

Analog-symbolic memory that tracks via reconsolidation

A fundamental part of a computational system is its memory, which is used to store and retrieve data. Classical computer memories rely on the static approach and are very different from human memories. Neural network memories are based on auto-associative attractor dynamics and thus provide a high level of pattern completion. However, they are not used in general computation since there are practically no algorithms to load an arbitrary landscape of attractors into them. In this sense neural network memory models cannot communicate well with symbolic and prior knowledge.We propose the design of a new memory based on localist attractor dynamics with reconsolidation called Reconsolidation Attractor Network (RAN). RAN combines symbolic and subsymbolic features in a very attractive way: it is based on attractors; enables pattern classification under missing data; and demonstrates dynamic reconsolidation, which is very useful for tracking changing concepts. The perception RAN enables is somewhat reminiscent of human perception due to its context sensitivity. Furthermore, it enables an immediate and clear interface with symbolic memories, including loading of attractors by means of trivial wiring, updating attractors, and retrieving them faster without waiting for full convergence. It also scales to any number of concepts. This provides a useful counterpoint to more conventional memory systems, such as random access memory and auto-associative neural networks.     

Editorial

Symbolic dynamics is a powerful tool in the study of dynamical systems. The purpose of symbolic dynamics is to provide a simplified picture of complicated dynamics, that gives some insight into its complexity. To this end, the state space of the system is partitioned in a finite number of pieces, and the exact trajectories of individual points are traded off by the trajectory relative to that partition. These so-called coarse-grained trajectories turn out to be realisations of a stationary random process with a finite alphabet. In particular, the entropy of a dynamical system can be approximated by the Shannon entropy of any of its symbolic dynamics (the finer the partition, the better the approximation). Today, symbolic dynamics is an independent field of theoretical physics and applied mathematics with applications to such important disciplines as cryptology, time series analysis, and data-compression.     

Symbolic dynamics of ventricular tachycardia and ventricular fibrillation

In this paper, the symbolic dynamics analysis was used to analyze the complexity of normal heartbeat signal (NSR), Ventricular tachycardia (VT) and ventricular fibrillation (VF) signals. By calculating the information entropy value of symbolic sequences, the complexities were quantified. Based on different information entropy values, NSR, VT and VF signals were distinguished with satisfactory results. The study showed that a sudden drop of symbolic sequence’s entropy value indicated that the patients most likely entered the episode of ventricular tachycardia and this was a crucial episode for the clinical treatment of patients. It had important clinical significance for the automatic diagnosis.