非高斯噪声激励下FitzHugh-Nagumo神经元系统的随机共振

研究了乘性非高斯噪声和加性高斯白噪声共同激励下FitzHugh-Nagumo(FHN) 神经元系统的随机共振问题. 利用路径积分法和两态模型理论, 推导出系统信噪比的表达式. 研究结果表明: 系统参数在不同的取值条件下, FHN神经元模型出现了随机共振和双重随机共振现象. 此外, 非高斯参数q在不同的取值条件下, 乘性噪声强度和加性噪声强度对信噪比的影响是不同的. 非高斯噪声的加入有利于增强FHN神经元系统的信号响应.     

非高斯噪声激励下FHN神经元系统的定态概率密度与平均首次穿越时间

研究了非高斯噪声激励下的FHN神经元系统,应用路径积分法和统一色噪声近似得到系统的定态概率密度函数和平均首次穿越时间表达式.发现了加性噪声强度Q能够诱导非平衡相变的产生,乘性噪声强度D、偏离参数p及关联时间τ₀均不能诱导非平衡相变发生;非高斯噪声的存在缩短了细胞神经元系统静息态和激发态之间的转化时间,有利于神经元信息的传递. 关键词： FHN神经元系统 非高斯噪声 定态概率密度 平均首次穿越时间     

关联高斯与非高斯噪声激励的FHN神经元系统的稳态分析

本文主要研究了关联乘性非高斯噪声和加性高斯白噪声共同激励的FHN(Fitz Hugh-Nagumo)神经元系统.利用路径积分法和统一色噪声近似,推导出该系统的定态概率密度函数表达式.通过研究发现,乘性噪声强度D、加性噪声强度Q、噪声自关联时间τ以及互关联系数λ均可以诱导系统产生非平衡相变现象,而非高斯参数q却不可以诱导系统产生非平衡相变现象.此外,我们还发现参数D和λ的增大有利于神经元系统从激发态向静息态转换,Q和τ的增大有利于神经元系统从静息态向激发态转换,q的增大会使得神经元系统停留在静息态的概率增加.     

非高斯噪声驱动下一维双稳系统的逻辑操作

本文以成功率作为逻辑随机共振的测度, 主要研究了非高斯噪声激励下一维双稳系统的逻辑随机共振现象, 并用平均首次通过时间的方法对此现象的机理进行了解释. 研究结果表明: 只有在适当的噪声强度带或关联时间带上, 成功率才会达到共振峰值. 通过对系统参数进行优化, 提高了系统实现逻辑操作的可靠性. 关键词： 逻辑随机共振 一维双稳系统 非高斯噪声 平均首次通过时间     

非高斯噪声驱动下非对称双稳系统的平均首次穿越时间与随机共振研究

研究了乘性非高斯噪声和加性高斯白噪声共同激励下非对称双稳系统的平均首次穿越时间和随机共振问题. 利用路径积分法和两态模型理论,推导出平均首次穿越时间和信噪比的表达式. 研究结果表明:势阱非对称性对两个不同方向的平均首次穿越时间的影响是不同的. 信噪比是加性噪声强度和势阱非对称性的非单调函数,系统出现了随机共振现象;信噪比是乘性噪声强度的单调函数,没有共振峰出现. 这说明该系统中乘性噪声强度和加性噪声强度对信噪比的影响是不同的. 关键词： 非高斯噪声 非对称双稳系统 平均首次穿越时间 随机共振     

色噪声激励下的FHN神经元系统

研究了色噪声激励下的FHN系统的相变问题和平均首次穿越时间,并推导了系统的定态概率密度函数和平均首次穿越时间表达式.结果表明,参数α,τ和β可以诱导相变,且存在锁定现象,各个参数对平均首次穿越时间都有很大影响,但影响效果有很大不同. 关键词： FHN系统 色噪声 平均首次穿越时间 定态概率密度     

简谐噪声激励下FitzHugh-Nagumo神经元的动力学行为

研究了简谐噪声激励下的FitzHugh-Nagumo神经元模型, 其放电形式、相干共振等动力学行为均受噪声阻尼参数和频率参数的影响.选择不同的参数所得到的神经元的放电形式不同.神经元存在共振特性,对某一频率的噪声有更强的响应,在此频率参数下的峰序列更有序,出现相干共振系数的极小值.噪声的阻尼参数越大,不同的频率成分越多,神经元的响应也变得杂乱,进而导致神经元与噪声的同步变弱,峰序列相干共振系数也相应增大. 关键词： 简谐噪声 FitzHugh-Nagumo神经元 相干共振 峰峰间隔     

简谐噪声激励下FitzHugh-Nagumo神经元的动力学行为

研究了简谐噪声激励下的FitzHugh-Nagumo神经元模型, 其放电形式、相干共振等动力学行为均受噪声阻尼参数和频率参数的影响.选择不同的参数所得到的神经元的放电形式不同.神经元存在共振特性,对某一频率的噪声有更强的响应,在此频率参数下的峰序列更有序,出现相干共振系数的极小值.噪声的阻尼参数越大,不同的频率成分越多,神经元的响应也变得杂乱,进而导致神经元与噪声的同步变弱,峰序列相干共振系数也相应增大.     

非周期二进制/M进制信号激励下非线性系统的非周期共振研究

研究单一非周期二进制或M进制信号激励下一类非线性系统的非周期共振现象及其度量方法,重点探讨了系统参数引起的非周期共振.提出了适用于非周期共振度量的响应幅值增益指标,并结合互相关系数和误码率展开研究.结果发现,互相关系数能够较好地描述系统输出和输入信号之间的同步性及波形相似性但无法刻画信号通过系统后被放大的程度.响应幅值增益能够较好地描述信号通过系统后幅值被放大的程度,但无法反映系统输出和输入信号之间的同步性及波形相似性.非周期共振发生在互相关系数取谷值和响应幅值增益取峰值处,且两种指标曲线反映的共振点相同.误码率在合适的阈值下可以描述系统输出和输入信号之间的同步性以及非周期信号通过系统后被放大的程度,误码率曲线可以直接给出非周期共振的共振区.单一非周期二进制或M进制信号激励下的非线性系统可以发生非周期共振,其共振效果需要综合互相关系数、响应幅值增益、误码率等指标进行度量.     

双色噪声激励下FHN神经元系统的稳态性质

应用统一色噪声理论研究了双色噪声激励下一维FitzHugh-Nagumo (FHN)神经元系统的动力学性质,即稳态概率分布函数和其平均值. 给出了FHN神经元系统的稳态概率密度和平均值的解析表达式. 结果表明: 乘性噪声的自关联时间τ ₁、加性噪声的自关联时间τ ₂、加性噪声强度α 和乘性噪声强度D都能够诱导非平衡相变的产生. α和D的增大有利于系统从激发态向静息态转换. τ₁, τ₂的增大有利于系统从静息态向激发态转换. 噪声强度和其自关联时间的作用完全相反.     

Effects of non-Gaussian noise on the dynamical properties of a logistic system

The dynamical properties of a tumor cell growth system described by the logistic system with coupling between non- Gaussian and Gaussian noise terms are investigated. The effects of the nonextensive index q on the stationary properties and the transient properties are discussed, respectively. The results show that the nonextensive index q can induce the tumor cell numbers to decrease greatly in the case of q 〉 1. Moreover, the switch from the steady stable state to the extinct state is speeded up as the increases of q, and the tumor cell numbers can be more obviously restrained for a large value of q. The numerical results are found to be in basic agreement with the theoretical predictions.     

Mean First Passage Time for System with Fluctuating Potential Barrier and Coupled Noise

In this paper we study the mean first passage time (MFPT) over a fluctuation potential barrier driven by a coupled noise. It is shown that the MFPT over the fluctuation potential barrier displays resonant activations as the function of the flipping rate of the fluctuation potential barrier, and as the function of the dichotomous noise transition rate.     

Mean first-passage time of an asymmetric bistable system driven by colour-correlated noise

In this paper, the effect of every parameter (including p, q, r, \la, \tau) on the mean first-passage time (MFPT) is investigated in an asymmetric bistable system driven by colour-correlated noise. The expression of MFPT has been obtained by applying the steepest-descent approximation. Numerical results show that (1) the intensity of multiplicative noise p and the intensity of additive noise q play different roles in the MFPT of the system, (2) suppression appears on the curve of the MFPT with small \la (e.g. \la<0.5) but there is a peak on the curve of the MFPT when \la is big (e.g. \la >0.5), and (3) with different values of r (e.g. r=0.1, 0.5, 1.5), the effort of \tau on the MFPT is diverse.     

Effect of time delay in FitzHugh-Nagumo neural model with correlations between multiplicative and additive noises

We study the effect of time delay in the FitzHugh-Nagumo neural model with correlations between multiplicative and additive noise terms. Based on the corresponding Fokker-Planck equation, the explicit expressions of the stationary probability distribution function (SPDF), the mean first passage time (MFPT) and the signal-to-noise ratio (SNR) are obtained, respectively. Research results show that: (i) the system undergoes a succession of two phase transitions (i.e., the reentrance phenomenon) as the noise correlation parameter is increased and a (single) phase transition as the time delay is increased. (ii) The MFPT as a function of the multiplicative noise intensity exhibits a maximum. This maximum for MFPT identifies the noise enhanced stability (NES) effect, the noise correlation parameter intensifies the NES effect while the time delay, and the additive noise intensity weakens it. (iii) The existence of a maximum in the SNR as a function of the multiplicative noise intensity is the identifying characteristic of the stochastic resonance (SR) phenomenon, the noise correlation parameter enhances the SR while the time delay, and the additive noise intensity weaken it.     

Stationary properties of a single-mode laser system with non-Gaussian and Gaussian noise

A single-mode laser system with non-Gaussian and Gaussian noise is investigated. The stationary mean value and the normalized variance of the laser intensity are numerically calculated under the condition that the stationary probability distribution function (SPDF) is derived. The SPDF as a function of the laser intensity exhibits a maximum. The maximum becomes smaller with the increase of the correlation intensity or the non-Gaussian parameter, where the later is a measure of the deviation from the Gaussian characteristic. The maximum becomes larger as the correlation time increases. The laser intensity stationary mean value decreases with the increase of the correlation intensity or the non-Gaussian parameter while increases with the correlation time increasing. The laser intensity normalized variance increases with the increase of the correlation intensity or the non-Gaussian parameter while decreases as the correlation time increases.