In this study,we developed a system based on deep space–time neural networks for gesture recognition.When users change or the number of gesture categories increases,the accuracy of gesture recognition decreases consi...In this study,we developed a system based on deep space–time neural networks for gesture recognition.When users change or the number of gesture categories increases,the accuracy of gesture recognition decreases considerably because most gesture recognition systems cannot accommodate both user differentiation and gesture diversity.To overcome the limitations of existing methods,we designed a onedimensional parallel long short-term memory–fully convolutional network(LSTM–FCN)model to extract gesture features of different dimensions.LSTM can learn complex time dynamic information,whereas FCN can predict gestures efficiently by extracting the deep,abstract features of gestures in the spatial dimension.In the experiment,50 types of gestures of five users were collected and evaluated.The experimental results demonstrate the effectiveness of this system and robustness to various gestures and individual changes.Statistical analysis of the recognition results indicated that an average accuracy of approximately 98.9% was achieved.展开更多
In recent years,space-division multiplexing(SDM)technology,which involves transmitting data information on multiple parallel channels for efficient capacity scaling,has been widely used in fiber and free-space optical...In recent years,space-division multiplexing(SDM)technology,which involves transmitting data information on multiple parallel channels for efficient capacity scaling,has been widely used in fiber and free-space optical communication sys-tems.To enable flexible data management and cope with the mixing between different channels,the integrated reconfig-urable optical processor is used for optical switching and mitigating the channel crosstalk.However,efficient online train-ing becomes intricate and challenging,particularly when dealing with a significant number of channels.Here we use the stochastic parallel gradient descent(SPGD)algorithm to configure the integrated optical processor,which has less com-putation than the traditional gradient descent(GD)algorithm.We design and fabricate a 6×6 on-chip optical processor on silicon platform to implement optical switching and descrambling assisted by the online training with the SPDG algorithm.Moreover,we apply the on-chip processor configured by the SPGD algorithm to optical communications for optical switching and efficiently mitigating the channel crosstalk in SDM systems.In comparison with the traditional GD al-gorithm,it is found that the SPGD algorithm features better performance especially when the scale of matrix is large,which means it has the potential to optimize large-scale optical matrix computation acceleration chips.展开更多
This paper describes the inverstigation devoted to establish suitable weights in a feed-forward neural network realizing the narrow-band filtering map in the case of adaptive line enhanccment(ALE) by the utility of th...This paper describes the inverstigation devoted to establish suitable weights in a feed-forward neural network realizing the narrow-band filtering map in the case of adaptive line enhanccment(ALE) by the utility of the optimum common learning rate back propagation (OCLR BP) algorithm. It is found that a feed-forward network with 64 linear input and output neurons, and 8 odd sigmoid neurons in the hidden layer, i.e. an (64→8→64) architecture, could establish the specific input-output function in the case of relatively low signal-to-noise radio. Only is an input signal consisting of mixed periodic and broad-band components available to the network system. After learning, both the 'fanning-in-connection patterns', each of which consists of weights fanning into a hidden-neuron from all the outputs of input-neurons, and the 'fanning-out-connection patterns', each of which consists of weights fanning out from a hidden-neuron to all the inputs of output-neurons, are tuned to the periodic signals. The nonlinear map formed by this neural network provided substantial improvement in performance over that formed by an Adaline-ALE with same frequency resolution.展开更多
针对短波信道中信号传输产生的码间串扰问题,文中结合人工神经网络(Artificial Neural Network,ANN)与递归最小二乘法(Recursive Least Square,RLS)算法,实现改进的RLS-ANN短波信道均衡算法。通过有监督与无监督分段均衡处理,实时校准...针对短波信道中信号传输产生的码间串扰问题,文中结合人工神经网络(Artificial Neural Network,ANN)与递归最小二乘法(Recursive Least Square,RLS)算法,实现改进的RLS-ANN短波信道均衡算法。通过有监督与无监督分段均衡处理,实时校准经由短波信道传播之后产生的信号幅度和相位失真。对实测的高频数据链(High Frequency Data Link,HFDL)短波信号进行均衡处理,结果表明RLS-ANN算法相比传统的最小均方算法(Least Mean Square,LMS)和RLS算法在星座图收敛速度、平均误差及误码率等方面效果更优,该算法通过降低误码率,可有效改善信号经由短波信道传输的通信质量。展开更多
设计一种应用于广播电视发射基站的信号实时信道均衡系统,该系统基于数字信号处理(Digital Signal Processing,DSP)算法,旨在提高信号传输质量并降低误码率。系统由信号预处理、信道估计、自适应均衡3个关键模块组成。信号预处理模块采...设计一种应用于广播电视发射基站的信号实时信道均衡系统,该系统基于数字信号处理(Digital Signal Processing,DSP)算法,旨在提高信号传输质量并降低误码率。系统由信号预处理、信道估计、自适应均衡3个关键模块组成。信号预处理模块采用自适应滤波技术去除噪声和干扰;信道估计模块利用频域分析技术精确估计信道参数;自适应均衡模块则通过最小均方误差(Least Mean Square,LMS)算法动态调整均衡器系数,以补偿信道失真。实验结果表明,该系统在城市、郊区、山区环境下均能显著提高信号质量,降低误码率,并提供足够的信道容量,满足广播电视信号的高质量实时传输需求。展开更多
To mitigate the linear and nonlinear distortions in communication systems, two novel nonlinear adaptive equalizers are proposed on the basis of the neural finite impulse response （FIR） filter, decision feedback arch...To mitigate the linear and nonlinear distortions in communication systems, two novel nonlinear adaptive equalizers are proposed on the basis of the neural finite impulse response （FIR） filter, decision feedback architecture and the characteristic of the Laguerre filter. They are neural FIR adaptive decision feedback equalizer （SNNDFE） and neural FIR adaptive Laguerre equalizer （LSNN）. Of these two equalizers, the latter is simple and with characteristics of both infinite impulse response （IIR） and FIR filters; it can use shorter memory length to obtain better performance. As confirmed by theoretical analysis, the novel LSNN equalizer is stable （0 〈α〈1）. Furthermore, simulation results show that the SNNDFE can get better equalized performance than SNN equalizer, while the latter exhibits better performance than others in terms of convergence speed, mean square error （MSE） and bit error rate （BER）. Therefore, it can reduce the input dimension and eliminate linear and nonlinear interference effectively. In addition, it is very suitable for hardware implementation due to its simple structure.展开更多
基金supported in part by the National Natural Science Foundation of China under Grant 61461013in part of the Natural Science Foundation of Guangxi Province under Grant 2018GXNSFAA281179in part of the Dean Project of Guangxi Key Laboratory of Wireless Broadband Communication and Signal Processing under Grant GXKL06160103.
文摘In this study,we developed a system based on deep space–time neural networks for gesture recognition.When users change or the number of gesture categories increases,the accuracy of gesture recognition decreases considerably because most gesture recognition systems cannot accommodate both user differentiation and gesture diversity.To overcome the limitations of existing methods,we designed a onedimensional parallel long short-term memory–fully convolutional network(LSTM–FCN)model to extract gesture features of different dimensions.LSTM can learn complex time dynamic information,whereas FCN can predict gestures efficiently by extracting the deep,abstract features of gestures in the spatial dimension.In the experiment,50 types of gestures of five users were collected and evaluated.The experimental results demonstrate the effectiveness of this system and robustness to various gestures and individual changes.Statistical analysis of the recognition results indicated that an average accuracy of approximately 98.9% was achieved.
基金supported by the National Natural Science Foundation of China(NSFC)(62125503,62261160388)the Natural Science Foundation of Hubei Province of China(2023AFA028)the Innovation Project of Optics Valley Laboratory(OVL2021BG004).
文摘In recent years,space-division multiplexing(SDM)technology,which involves transmitting data information on multiple parallel channels for efficient capacity scaling,has been widely used in fiber and free-space optical communication sys-tems.To enable flexible data management and cope with the mixing between different channels,the integrated reconfig-urable optical processor is used for optical switching and mitigating the channel crosstalk.However,efficient online train-ing becomes intricate and challenging,particularly when dealing with a significant number of channels.Here we use the stochastic parallel gradient descent(SPGD)algorithm to configure the integrated optical processor,which has less com-putation than the traditional gradient descent(GD)algorithm.We design and fabricate a 6×6 on-chip optical processor on silicon platform to implement optical switching and descrambling assisted by the online training with the SPDG algorithm.Moreover,we apply the on-chip processor configured by the SPGD algorithm to optical communications for optical switching and efficiently mitigating the channel crosstalk in SDM systems.In comparison with the traditional GD al-gorithm,it is found that the SPGD algorithm features better performance especially when the scale of matrix is large,which means it has the potential to optimize large-scale optical matrix computation acceleration chips.
文摘This paper describes the inverstigation devoted to establish suitable weights in a feed-forward neural network realizing the narrow-band filtering map in the case of adaptive line enhanccment(ALE) by the utility of the optimum common learning rate back propagation (OCLR BP) algorithm. It is found that a feed-forward network with 64 linear input and output neurons, and 8 odd sigmoid neurons in the hidden layer, i.e. an (64→8→64) architecture, could establish the specific input-output function in the case of relatively low signal-to-noise radio. Only is an input signal consisting of mixed periodic and broad-band components available to the network system. After learning, both the 'fanning-in-connection patterns', each of which consists of weights fanning into a hidden-neuron from all the outputs of input-neurons, and the 'fanning-out-connection patterns', each of which consists of weights fanning out from a hidden-neuron to all the inputs of output-neurons, are tuned to the periodic signals. The nonlinear map formed by this neural network provided substantial improvement in performance over that formed by an Adaline-ALE with same frequency resolution.
文摘针对短波信道中信号传输产生的码间串扰问题,文中结合人工神经网络(Artificial Neural Network,ANN)与递归最小二乘法(Recursive Least Square,RLS)算法,实现改进的RLS-ANN短波信道均衡算法。通过有监督与无监督分段均衡处理,实时校准经由短波信道传播之后产生的信号幅度和相位失真。对实测的高频数据链(High Frequency Data Link,HFDL)短波信号进行均衡处理,结果表明RLS-ANN算法相比传统的最小均方算法(Least Mean Square,LMS)和RLS算法在星座图收敛速度、平均误差及误码率等方面效果更优,该算法通过降低误码率,可有效改善信号经由短波信道传输的通信质量。
文摘设计一种应用于广播电视发射基站的信号实时信道均衡系统,该系统基于数字信号处理(Digital Signal Processing,DSP)算法,旨在提高信号传输质量并降低误码率。系统由信号预处理、信道估计、自适应均衡3个关键模块组成。信号预处理模块采用自适应滤波技术去除噪声和干扰;信道估计模块利用频域分析技术精确估计信道参数;自适应均衡模块则通过最小均方误差(Least Mean Square,LMS)算法动态调整均衡器系数,以补偿信道失真。实验结果表明,该系统在城市、郊区、山区环境下均能显著提高信号质量,降低误码率,并提供足够的信道容量,满足广播电视信号的高质量实时传输需求。
基金Supported partially by the National Natural Science Foundation of China (Grant No. 60971104)the Program for New Century Excellent Talents in University of China (Grant No. NCET-05-0794)the Doctoral Innovation Fund of Southwest Jiaotong University
文摘To mitigate the linear and nonlinear distortions in communication systems, two novel nonlinear adaptive equalizers are proposed on the basis of the neural finite impulse response （FIR） filter, decision feedback architecture and the characteristic of the Laguerre filter. They are neural FIR adaptive decision feedback equalizer （SNNDFE） and neural FIR adaptive Laguerre equalizer （LSNN）. Of these two equalizers, the latter is simple and with characteristics of both infinite impulse response （IIR） and FIR filters; it can use shorter memory length to obtain better performance. As confirmed by theoretical analysis, the novel LSNN equalizer is stable （0 〈α〈1）. Furthermore, simulation results show that the SNNDFE can get better equalized performance than SNN equalizer, while the latter exhibits better performance than others in terms of convergence speed, mean square error （MSE） and bit error rate （BER）. Therefore, it can reduce the input dimension and eliminate linear and nonlinear interference effectively. In addition, it is very suitable for hardware implementation due to its simple structure.
