Method for processing input/output signals in dynamic cognitive neural function reconstruction
An input-output and neural function technology, applied in the field of multi-input-output dynamic system signal processing, can solve problems such as input-output signal processing methods with no convergence speed
Active Publication Date: 2016-11-09
SICHUAN DONGDING LIZHI INFORMATION TECH CO LTD
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] However, for systems with dynamic cognitive neural networks, approximation and recursion methods are usually used (for example, the feedforward neural approximation network included in the prior art), and there is currently no input-output signal processing method with a faster convergence speed ( See Carvalho J P. Rule Based Fuzzy Cognitive Maps-Qualitative Dy-namics [OL]. http: / / www.google.com)
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0051] like figure 1 As shown, according to a preferred embodiment of the present invention, the present invention provides a method for processing input and output signals in dynamic cognitive neural function reconstruction, including:
[0052] (1) Construct the detection signal reception matrix according to the state of the dynamic cognitive neural network;
[0053] (2) Singular value decomposition is carried out to the detection signal reception matrix at the current moment;
[0054] (3) Construct a feedback signal matrix according to the detection signal receiving matrix at the next moment and the singular value decomposition result;
[0055] (4) Amplify the signal and iterate the dynamic cognitive neural network.
[0056] Preferably, said step (1) includes:
[0057] (11) According to the state of the dynamic cognitive neural network, the state at time t, obtain the time t to be input to node N i The effective signal S i(t) and input to node N at time t+1 i The effec...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
In order to the speed of processing node signals in a dynamic cognitive neural network, the invention provides a method for processing input / output signals in dynamic cognitive neural function reconstruction. The method comprises the following steps: constructing a detection signal reception matrix according to the state of the dynamic cognitive neural network; carrying out singular value decomposition on the detection signal reception matrix at the current moment; constructing a feedback signal matrix according to the detection signal reception matrix at the next moment and the singular value decomposition result; and amplifying the signals and iterating the dynamic cognitive neural network. According to the method provided by the invention, transformation is carried out on the basis of proportion, between signal characteristics at different moments, of nodes, and signal screening is carried out by utilizing phase stability of the input signal to be input to a multi-input multi-output dynamic system, so that the modeling convergence speed is improved and then the speed of processing the input / output signals of each node is improved.
Description
technical field [0001] The present invention relates to the technical field of multi-input-output dynamic system signal processing, and more specifically, to an input-output signal processing method in dynamic cognitive neural function reconstruction. Background technique [0002] Due to the strong function approximation ability of artificial neural network, it can fully approximate any continuous nonlinear function with arbitrary precision, and has self-adaptive and self-learning ability for complex uncertain problems. , biological image recognition, mechanical and electrical equipment design, optimization calculation and other fields provide a new efficient data processing method. [0003] However, for systems with dynamic cognitive neural networks, approximation and recursion methods are usually used (for example, the feedforward neural approximation network included in the prior art), and there is currently no input-output signal processing method with a faster convergen...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06N3/06
CPCG06N3/06
Inventor 周琳陈林瑞
Owner SICHUAN DONGDING LIZHI INFORMATION TECH CO LTD