Hopfield neural network efficiency in multicast routing

What is it about?

A neural network approach is presented to solve the delay constraint and loss data multicast routing problem, an NP-complete problem. The multicast tree is obtained in line with optimal (minimum) routing costs which are subjected to integrated delay and loss data constraints. The proposed routing algorithm is designed to find the optimal (shortest) path taking into account traffic conditions (the incoming traffic flow, routers occupancy, and link capacities), avoiding packet loss data to the input buffer overflow and combine them into the union. The experimental results showed that the proposed method using the discrete-time equation has achieved a significant difference in iterations and execution time compared to other proposals such as Runge–Kutta, Euler equations, ant colony algorithm and Hierarchical Hopfield neural network which contributes to improving the performance of Internet of Things networks in terms of the durability of interconnection between devices. In accordance with the extent of the impact of quality of service restrictions on the network, and the maintenance of the cost-effectiveness of the route, the algorithm achieved a decrease in execution time ranging from 27% to 81%, in addition to a decrease in the number of iterations by 44% to 66%.

Featured Image

Photo by Google DeepMind on Unsplash

Photo by Google DeepMind on Unsplash

Why is it important?

The field of communication is known as providing quality of service restrictions according to each interface node to increase the applications which are led to the need to provide a network that can accommodate data packets to exchange between applications without losing those packets. The energy saving factor is important to show how efficiently applications work, in addition to the importance of the energy factor with the spread of mobile devices that carry Internet of Things applications to control home automation systems and other applications, especially the delay factor, which expresses the speed of the response of the application, which has increased in importance with the increasing use of networks of broadcast devices, tracking systems and wireless sensors in the Internet of Things. One of the constrains that we may face for these applications is cost, time and jitter which is known as data loss rate, and have an impact on network resources. The work targets the problem of multicast routing of communication networks in light of the massive development of the Internet of Things (IoT), which refers to the problem of enabling a huge number of IoT devices to achieve their quality of service limitations in the next generation networks. This work uses artificial intelligence (AI) techniques to solve this major problem.

Perspectives