Vertical switching algorithm based on interval type-2 fuzzy neural network

Abstract

The invention relates to a vertical switching algorithm based on an interval type-2 fuzzy neural network, which belongs to the technical field of mobile communication and specifically comprises the following steps of: firstly, defining a historical access rate in a network pre-selection stage, and setting a threshold value in combination with the number of current candidate network sets; accordingto the received signal strength and the remaining available bandwidth, preliminarily selecting all networks in a user receiving range; and in a vertical switching determination stage, using the timedelays, the packet loss rates and the bit error rates of the remaining candidate networks as the inputs of the interval type-2 fuzzy neural network, completing fuzzy logic reasoning by using the structure of a feedforward neural network, and calculating an output decision value after training, so that the optimal access network is selected. Finally, an experimental result shows that the algorithmcan effectively reduce the error probability of the switching decision, reduce the switching failure and the switching frequency and improve the total throughput of the network while ensuring lower time complexity.

Description

technical field [0001] The invention belongs to a vertical switching algorithm in an ultra-dense heterogeneous wireless network formed by a heterogeneous wireless cellular network and a wireless local area network of 5G macro base stations and micro base stations, and belongs to the field of mobile communication. In particular, it relates to an algorithm for vertical switching decision using interval type two fuzzy neural network system. Background technique [0002] The increase in the application of smart devices and information and communication technology has accelerated the evolution of wireless technology to 5G networks. In order to solve the problems caused by the rapid growth of these devices and data traffic, macro, micro base stations and other wireless technologies are used in new network scenarios Heterogeneous ultra-dense networking architecture for deployment. This ultra-dense deployment sets up a large number of micro base stations within the scope of the mac...

AI Technical Summary

Problems solved by technology

[0005] In the current ultra-dense heterogeneous wireless network environment, the traditional vertical handover algorithm does not express the fuzziness and randomness of the network at the same time, and only considers the fuzziness of vertical handover, and even if there are studies using type-2 fuzzy logic, it is relatively On the surface, the subjectivity of fuzzy logic itself is not considered, because the membership functions and rules are artificially set, and this paper uses neural networks to train the parameters of type II fuzzy logic, making the entire system decision-making more in line with real network scenarios

Images