Abstract
. The article discusses a neural network in the form of self-organizing Kohonen maps (SOKM) as a method of structuring the service signals of the standard IEEE 802.22. Examined the neural network of the SOKM as a cognitive level classification signaling standard IEEE 802.22. It was built simulation model considered in the development environment MATLAB. Implemented training Hebb rule is defined as a training sample 8 randomly-generated sequences. Hebb rule change provides connections in one direction only, which in this case could lead to a wrong determination of the values of the weights. To avoid this, the concept entered by smoothing function. The complexity of the input of the space was taken in a simplified way due to the limited amount of computing resources. To simulate a real input sequence should review the amount of the input stream and determine the appropriate computing resources for data flows. The result of this work showed that the SOKM is able to cluster complex signals, in turn solves the problem, which was staged.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call