Spectrum accessing optimization in congestion times in radio cognitive networks based on chaotic neural networks

TL;DR

This paper proposes a chaotic recurrent neural network approach to optimize spectrum access during congestion in cognitive radio networks, aiming to reduce interference and improve throughput.

Contribution

It introduces a novel channel assignment method using chaotic neural networks that adapts to environmental responses and reduces interference in spectrum sharing.

Findings

Reduced interference with chaotic neural network-based access

Increased throughput during congestion times

Lower operational costs for secondary users

Abstract

Based on the theory of the Federal Communications Commission, the spectrum available on cognitive radio networks is limit and the non-optimal use of the spectrum necessitates the need for a telecommunications model, so that this pattern can exploit the existing spectral positions. In this spectrum subscription scenario, when the primary users are not present, it is also possible to assign this telecommunication to tenants who are unauthorized or secondary. The challenge of using this scenario is to allocate time-frequency resources to them and how to access nodes in one channel without any interactions between primary and secondary users and the throughput will increase. The main idea of this research is using chaotic recurrent neural network for improving access to spectrum in congestion times and the main purposes are reduce interference and increase throughput in cognitive radio networks. In this method, in addition to the throughput, the amount of unwanted blockage of packets, the reduction of the cost of operations for secondary users, the hardware requirements for secondary users and the coefficient of justice are considered which in fact, it is a new channel assignment process with respect to the environment response, the updates the probability that the channels are empty in subsequent periods, and increases the permeability by reducing interference with chaotic recurrent neural network.