一种基于自组织神经网络的语音识别系统

建立了一种基于自组织神经网络的语音识别系统。对语音信号进行了预处理,提取了语音信号的线性预测系数、线性预测倒谱系数和Mel倒谱特征系数,建立了基于自组织神经网络的识别判决模型。深入分析和改进了自组织神经网络的分类聚类能力,通过加强训练和设定阈值函数的方法,有效地确定了边界神经元的归属,划分出了合理的输出模式类。验证了自组织神经网络适合于处理孤立词语音识别,并具有快速性和结构简单等特征。MATLAB仿真实验表明,语音识别率达到96%。     

基于最大流的无线mesh网络负载均衡信道分配算法

针对无线mesh网络中各信道间的干扰情况及流量负载均衡问题,定义了负载均衡的目标优化函数,设计一种基于最大流的负载均衡信道分配算法。该算法优先考虑为负载量比较大的链路分配更多的带宽,使得一些关键链路能够承受较大的流量负载,实现各信道的负载均衡,减少网络拥塞,降低分组丢失率和延迟。实验结果表明,该算法较好地平衡了网络负载,在网络业务较重的情况下还能获得较好的网络性能。     

Depth video spatial and temporal correlation enhancement algorithm based on just noticeable rendering distortion model

Spatial and temporal inconsistency of depth video deteriorates encoding efficiency in three dimensional video systems. A depth video processing algorithm based on human perception is presented. Firstly, a just noticeable rendering distortion (JNRD) model is formulated by combining the analyses of the influence of depth distortion on virtual view rendering with human visual perception characteristics. Then, depth video is processed based on the JNRD model from two aspects, spatial and temporal correlation enhancement. During the process of spatial correlation enhancement, depth video is segmented into edge, foreground, and background regions, and smoothened by Gaussian and mean filters. The operations of the temporal correlation enhancement include temporal–spatial transpose (TST), temporal smoothing filter and inverse TST. Finally, encoding and virtual view rendering experiments are conducted to evaluate the proposed algorithm. Experimental results show that the proposed algorithm can greatly reduce the bit rate while it maintains the quality of virtual view.     

Investigations on the nanostructures of GaN,InN and InxGa1−xN

A controllable approach to the formation of III- nitride nanocrystalline structures using hydrothermal assisted method is presented. The structural and morphological properties of the prepared nanostructures are analyzed using X-ray diffraction, Fast Fourier Transformation and transmission electron microscope techniques. The temperature dependent structural formation of nitride nanostructures have been systematically investigated using X-ray diffraction. Raman spectra of the samples grown at optimized condition exhibited different phonon modes of the respective nitrides (GaN, InN and In_xGa_1−xN). Nanoparticles and nanorods formation of the indium nitride and indium gallium nitride are observed in the TEM micrographs. FFT analysis revealed that the synthesized III-nitride nanostructures are of good crystalline quality. Nanorods of these nitrides showed better crystalline quality than the nanoparticles in the FFT reflections.     

PWCE design with Hop-Limited p-Cycles for dynamical protection provision in WDM networks

In future high-capacity wavelength division multiplexed (WDM) optical networks; the failure of a network component such as a fiber link can lead to severe disruption in the networks' traffic. Hence, it is imperatively important to provide fast and 100% protection in WDM optical networks. In this paper, a Hop-Limited p-Cycle-based approach is employed for protected working capacity envelope (PWCE) design. The results reveal the cycle hop not only affects the total protected working units but also the blocking probability performance.     

Low-latency neighbor discovery algorithm based on multi-beacon message in mobile low-duty-cycle sensor network

Neighbor discovery enables nodes in the networks to discover each other through simple information interaction,which was suitable for the new mobile low duty cycle sensor network (MLDC-WSN).However,because the nodes in MLDC-WSN can move randomly and sleep,the network topology was changed frequently,which results in that some nodes need a lot of energy and time to find their neighbors.How to realize fast neighbor discovery for all nodes in the network was a difficult problem in current research.To solve this problem,a new low-latency neighbor discovery algorithm based on multi-beacon messages was proposed.In this algorithm,the nodes were discovered by sending a short beacon message through their neighbor nodes,and by adjusting the time and frequency of beacon message sent,a lower neighbor discovery delay was obtained.Eventually,through quantitative analysis and simulation experiments,it is found that compared with existing algorithms,this algorithm can find all neighbor nodes in MLDC-WSN with less energy consumption,lower latency and greater probability.     

A self-adaptive weighted affinity propagation clustering for key frames extraction on human action recognition

In this paper, we propose a novel approach for key frames extraction on human action recognition from 3D video sequences. To represent human actions, an Energy Feature (EF), combining kinetic energy and potential energy, is extracted from 3D video sequences. A Self-adaptive Weighted Affinity Propagation (SWAP) algorithm is then proposed to extract the key frames. Finally, we employ SVM to recognize human actions on the EFs of selected key frames. The experiments show the information including whole action course can be effectively extracted by our method, and we obtain good recognition performance without losing classification accuracy. Moreover, the recognition speed is greatly improved.     

基于以太网技术的音视频信号切换器的设计

提出一种带以太网接口的智能音视频切换器。该系统以Mega64单片机为核心,实现了4×1的音频、视频信号监测和自动切换输出。通过以太网接口连接监控中心,实现远程参数配置和监控功能。同时,该系统在无上位机时,可独立工作,灵活性较好。     

Super-Stretchable,Anti-Freezing,Anti-Drying Organogel Ionic Conductor for Multi-Mode Flexible Electronics

Due to their intrinsic flexibility, tunable conductivity, multiple stimulus-response, and self-healing ability, ionic conductive hydrogels have drawn significant attention in flexible/wearable electronics. However, challenges remain because traditional hydrogels inevitably faced the problems of losing flexibility and conductivity because of the inner water loss when exposed to the ambient environment. Besides, the water inside the hydrogel will freeze at the water icing temperatures, making the device hard and fragile. As a promising alternative, organogels have attracted wide attention because they can, to some extent, overcome the above drawbacks. Herein, a kind of organogel ionic conductor (MOIC) by a self-polymerization reaction is involved, which is super stretchable, anti-drying, and anti-freezing. Meanwhile, it can still maintain high mechanical stability after alternately loading/unloading at the strain of 600% for 600 s (1800 cycles). Using this MOIC, high-performance triboelectric nanogenerator (TENG) is constructed (MOIC-TENG) to harvest small mechanical energy even the MOIC electrode underwent an extremely low temperature. In addition, multifunctional flexible/wearable sensors (strain sensor, piezoresistive sensor, and tactile sensor) are realized to monitor human motions in real time, and recognize different materials by triboelectric effect. This study demonstrates a promising candidate material for flexible/wearable electronics such as electronic skin, flexible sensors, and human-machine interfaces.