Impact of delays and rewiring on the dynamics of small-world neuronal networks with two types of coupling

We study synchronization transitions and pattern formation on small-world networks consisting of Morris-Lecar excitable neurons in dependence on the information transmission delay and the rewiring probability. In addition, networks formed via gap junctional connections and coupling via chemical synapses are considered separately. For gap-junctionally coupled networks we show that short delays can induce zigzag fronts of excitations, whereas long delays can further detriment synchronization due to a dynamic clustering anti-phase synchronization transition. For the synaptically coupled networks, on the other hand, we find that the clustering anti-phase synchronization can appear as a direct consequence of the prolongation of information transmission delay, without being accompanied by zigzag excitatory fronts. Irrespective of the coupling type, however, we show that an appropriate small-world topology can always restore synchronized activity if only the information transmission delays are short or moderate at most. Long information transmission delays always evoke anti-phase synchronization and clustering, in which case the fine-tuning of the network topology fails to restore the synchronization of neuronal activity.     

Double image encryption by using Arnold transform and discrete fractional angular transform

Based on Arnold transform and discrete fractional angular transform, a double image encryption algorithm is designed. Two original images are regarded as the amplitude and phase of a complex function. Arnold transform is introduced for scrambling the pixels at a local area of the complex function. Subsequently the changed complex function is converted by discrete fractional angular transform. The operations mentioned will be performed many times. The amplitude of final output complex function is the encrypted image and its phase is regarded as the key of encryption algorithm. The parameters of the two transforms serve as the additional keys for enhancing the security. Some numerical simulations have been done to validate the performance of this encryption scheme.     

低偏振相关损耗全息光栅波分复用器的设计

对基于掺杂铌酸锂材料的体光栅波分复用器主要结构和光栅记录参量提出了一种优化设计方法.采用严格耦合波理论,对两种偏振不同入射角和不同记录晶体厚度下的器件关键性能参量,如插入损耗和偏振相关损耗等进行了优化.数值结果证明采用优化设计的晶体厚度和光栅记录时的入射角,在获得相当低的偏振相关损耗的同时,也能够获得较低的插入损耗,实现了综合性能优化的波分复用器设计.实验结果证明用优化设计的参量能有效降低波分复用器件的插入损耗和偏振相关损耗.     

Statistical thermodynamic formalism in the solution of information theory problems

This is the first of several papers dealing with the application of statistical thermodynamic methodology to the solution of coding and communication theory problems. Emphasis is placed on the various ensemble techniques of statistical mechanics, the words or samples of a message taking the place of molecules in the prototype physical system. Analogs of temperature, internal energy, pressure, chemical potential, volume, entropy, etc., are developed. The isomorphism with thermodynamics is complete and these quantities transform (for example, by partial differentiation) in exactly the same way as the prototype physical quantities. The methods are nicely applicable to coding cases involving sources with memory, in which case, correlation can be discussed in terms of analog coupling energies between signals or words so that the store of many-body-problem techniques can be used. In addition, the manipulative freedom stemming from the possibility of choosing from a multiplicity of ensembles constrained by intensive parameters proves a distinct advantage. A concrete example dealing with the choice of a compact code for a nonextended source with memory is presented. The compact code is derived, and some discussion is given concerning the breadth of its power spectrum. In a following paper, its autocorrelation function within the framework of pulse code modulation is derived and transformed by Wiener theory so that the power spectrum is directly exhibited (along with the spectra for several other cases).Research supported under AFOSR Grant No. 70-1877. The present work is contribution No. 2643 of the Department of Chemistry, University of California-Los Angeles.     

Research on Multi-Terminal’s AC Offloading Scheme and Multi-Server’s AC Selection Scheme in IoT

Mobile Edge Computing (MEC) technology and Simultaneous Wireless Information and Power Transfer (SWIPT) technology are important ones to improve the computing rate and the sustainability of devices in the Internet of things (IoT). However, the system models of most relevant papers only considered multi-terminal, excluding multi-server. Therefore, this paper aims at the scenario of IoT with multi-terminal, multi-server and multi-relay, in which can optimize the computing rate and computing cost by using deep reinforcement learning (DRL) algorithm. Firstly, the formulas of computing rate and computing cost in proposed scenario are derived. Secondly, by introducing the modified Actor-Critic (AC) algorithm and convex optimization algorithm, we get the offloading scheme and time allocation that maximize the computing rate. Finally, the selection scheme of minimizing the computing cost is obtained by AC algorithm. The simulation results verify the theoretical analysis. The algorithm proposed in this paper not only achieves a near-optimal computing rate and computing cost while significantly reducing the program execution delay, but also makes full use of the energy collected by the SWIPT technology to improve energy utilization.     

树上秘密共享体制的信息率

研究以树G为通道结构的秘密共享体制的最优信息率ρ（G）。得到了ρ（G）＝2／3的要条件。证明了ρ（G）不会介于实数区间（3／5，2／3）中，给出了以树G的阶数表示的ρ（G）的下界，求出两类具有某种结构的树的最优信息率。     

基于主元分析与动态时间弯曲的故障诊断方法及应用研究

轴承是工程实际中常用而又极易损坏的部件,特别是对其早期微弱响应的辨识,具有重要的社会价值和意义。为提高运转轴承的安全可靠性和可维护性,提出了基于主元分析与动态时间弯曲距离的故障诊断方法,它可以准确对早期微弱动态响应辨识、诊断。该方法首先将典型故障样本信号与待测信号小波去噪并EMD分解,并对若干固有模态分量主元分析求取主元,然后对主元分量进行分析,获得相关特征值组成特征向量,计算待测信号与已知故障样本信号特征向量的弯曲距离,弯曲距离越小表明两信号越相似,从而辨识故障。此外,还可将其应用于转子、碰磨、齿轮故障诊断中,工程应用实例表明该方法可以准确故障分类,高效故障诊断。     

Preparation and photochromic properties of pyrazolones/polyvinyl alcohol composite films

Novel photochromic composite films have been successfully fabricated by dispersing pyrazolone derivative:1,3-Diphenyl-4-(3-chlorobenzal)-5-hydroxypyrazole 4-phenylsemicarbazone (1a) into hydrosol of polyvinyl alcohol (PVA). The microstructure, photochromic behaviors and thermal bleaching properties were investigated by Raman spectroscopy, X-ray powder diffraction (XRD), field emission scanning electron microscopy (FE-SEM) and ultraviolet-visible absorption spectroscopy (UV-vis). The results showed that 1a was not only blended but also well dispersed in the PVA polymer films with a suitable content of chromophore. Upon UV light irradiation, the composite films gradually changed from colorless to yellow and recovered fully to the initial state upon thermal bleaching. The time constants of photochromic reactions were almost the same as those of 1a observed in their crystalline state, indicating that the photochromic phenomenon is barely disturbed by the polymer matrix.     

连续值域信息系统的规则提取与知识约简

研究连续值域信息系统的规则提取和知识约简，给出规则提取的完整方法和2种新的知识约简的概念，讨论各种协调集之间的关系，探讨系统允许误差变化对规则提取的影响。     

一种新的思维模型:多节拍主从结构离散神经网络(英)

本文基于McCulloch和Pitts提出的二值神经元阈值模型，以Hopfield联想神经网络为核心，加上两个从网络，构造了一个由节拍各异的三个离散网络组成的神经网络系统．本文分析了该系统的稳定性，并定性讨论了其联想思维功能．