SPATIO-TEMPORAL PATTERN RECOGNITION USING A SPIKING NEURAL NETWORK AND PROCESSING THEREOF ON A PORTABLE AND/OR DISTRIBUTED COMPUTER

摘要

A kind of system and method, for characterizing pattern, wherein spike neural network is attacked petite person's layer neuron and is provided. The spike neural network has the neuron of transmission signal connection between the neuron of multiple connections. Spiking neuron is induced to be designated in model. The neuron and the global regulation unit of signal are transmitted between each neuron and the global regulation unit. Thus neuron transmits a signal at least one other neuron to each neuron at least one other neuron, these neurons and at least one other neuron come identical layer. Each neuron of spike is synchronous according to many active neuronals and neuron. Airtight petite person's pattern submits to spike spike neural network of the spike neural network for formation sequence, and sequence spike (I) is in modulation and as time synchronization spike and (II) are by adjusting the global regulation unit. At least one pattern is characterized in that the spike neural network generated according to sequence spikes.