Title: A learning-based approach to improving multicast network performance

Authors: Hazem H. Abdulmajeed; Hesham A. Hefny; Assem Alsawy

Addresses: Faculty of Studies for Statistical Research (FSSR), Cairo University, Cairo, Egypt ' Faculty of Studies for Statistical Research (FSSR), Cairo University, Cairo, Egypt ' Faculty of Computer Science and Information Technology, Al-Ahram Canadian University, Giza, Egypt

Abstract: A neural network approach is recommended by us in this research paper, to solve the problem of multicast routing subject to some quality of service (QoS) restrictions in communications and internet of things (IoT) networks, which is considered a complete nondeterministic polynomial (NP) problem. This approach was taken to identify a multicast tree that satisfies those restrictions, in particular cost, delay, and data loss rate. The exemplary (shortest) path is identified by the recommended routing algorithm considering the traffic conditions (the incoming traffic flow, routers occupancy, and link capacities). The experimental results showed a significant difference in obtaining the exemplary path that was executed by the recommended method using the Hopfield neural network (HNN) approach, besides the number of iterations. Furthermore, the execution time is less compared with the recommendations of heuristic algorithms, such as the ant colony optimisation algorithm (ACO) and genetic algorithms (GAs).

Keywords: multicast routing; internet of things; IoT; neural network; quality of services; QoS; heuristic algorithms.

DOI: 10.1504/IJCNDS.2023.133901

International Journal of Communication Networks and Distributed Systems, 2023 Vol.29 No.6, pp.631 - 652

Received: 26 Jul 2022
Accepted: 21 Sep 2022
Published online: 05 Oct 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article