Signal processing method and system, spiking neural network and neuron model

• Main Authors: QU HUANYU, ZHANG WEIHAO
• Format: Patent
• Language: Chinese; English
• Published: 04.10.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention provides a signal processing method and system, a spiking neural network and a neuron model, and belongs to the technical field of computers, the method is applied to the neuron model in the spiking neural network, the spiking neural network comprises a plurality of neuron models, and the method comprises the following steps: when a received pulse signal comprises a reset signal, resetting the reset signal; and setting the state of the neuron model as a specified state. According to the embodiment of the invention, the state of the neuron module can be self-adaptively reset, so that the processing capability of the spiking neural network in the time dimension is improved. 本公开提供了一种信号处理方法及系统、脉冲神经网络、神经元模型，属于计算机技术领域，该方法应用于脉冲神经网络中的神经元模型，脉冲神经网络包括多个神经元模型，方法包括：在接收到的脉冲信号包括重置信号的情况下，将神经元模型的状态设置为指定状态。根据本公开的实施例能够自适应地重置神经元模块的状态，从而提升脉冲神经网络在时间维度的处理能力。