SOFTWARE IMPLEMENTATION OF A HYBRID FIS–NN MODEL FOR NETWORK CHANNEL STATE ASSESSMENT USING TELEMETRY DATA

Read full text Summarise with VIRAT
Authors: Alexey Andreyevich Gololobov
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This article presents a software framework for assessing the state of a network communication channel using a hybrid FIS–NN model that combines fuzzy inference with a multilayer neural network. The developed prototype implements a complete processing pipeline for telemetry data, including log import, traffic parameter preprocessing, initial evaluation via a fuzzy inference system, and subsequent neural-network-based refinement. The software architecture comprises modules for data import, normalization, hybrid evaluation, conditional route selection, and visualization. Experimental studies conducted under various load scenarios confirm the robustness of the hybrid model to noise disturbances and demonstrate improved classification accuracy compared to threshold-based approaches. The proposed software prototype can serve as a foundation for integration into operational monitoring systems and QoS management frameworks in modern telecommunication networks.

Fields of Study & Subjects

Network Traffic and Congestion Control Software-Defined Networks and 5G Fuzzy Logic and Control Systems

Publication details

Journal / source
Infokommunikacionnye tehnologii
Volume
23
Issue
4
Pages
17–17
Publication date
6 March 2026
Language
EN