Investigation of Routing Efficiency in Self-Organized Networks Based on Swarm Intelligence
Students Name: Butchak Ostap Romanovych
Qualification Level: magister
Speciality: Information Communication Networks
Institute: Institute of Telecommunications, Radioelectronics and Electronic Engineering
Mode of Study: full
Academic Year: 2021-2022 н.р.
Language of Defence: ukrainian
Abstract: The high rate of proliferation of unmanned aerial vehicles (UAVs) next to the dynamic development of wireless communication technologies has contributed to the emergence of a new type of communication network - self-organizing UAV – Flying Ad Hoc Networks(FANET) [1]. The main features of self-organizing UAVnetworks are the high level of mobility, frequent topology changes and large distances between nodes, compared with networks based on mobile devices (MANET) and vehicles (VANET) [2, 3]. To date, there is no universal protocol that takes into account all the features of self-organizing UAV networks. Much work has been done in recent years to implement swarm intelligence (SI) methods and create adaptive routing protocols for self-organizing networks [4, 5]. Traditional, centralized approaches lack scalability and fault tolerance, SI methods provide natural solutions through distributed adaptive routing approaches for self-organizing networks. The analysis of research and publications in the field of swarm intelligence, is one of the sections of artificial intelligence, has proved the promise of applying ant and bee optimization algorithms to solve the routing problem in self-organizing networks. Stochastic SI -based routing algorithms dominate over most deterministic classical routing algorithms. The bee colony method is one of the newest and promising multi-agent intelligent optimization methods and has proven itself in various fields, including telecommunications when solving the routing problem in self-organizing networks. Thus, improving the efficiency of routing in self-organizing UAV networks based on Swarm Intelligence is an urgent task of great importance for the development of telecommunications. The object of the study is self-organizing FANET networks. Subject of research - routing algorithms for FANET networks. The aim of the master’s qualification work is to improve the efficiency of traffic routing in self-organizing networks of drones based on swarm intelligence. In order to achieve the goal, the work consistently solves the following tasks: 1. Analysis of the state and prospects of development of self-organizing networks of unmanned aerial vehicles. 2. Analysis of features of functioning of routing algorithms based on a bee colony for self-organizing networks. 3. Development of a bee colony based routing algorithm taking into account the requirements and peculiarities of self-organizing UAV networks. 4. Conduct research to assess the effectiveness of the proposed algorithm by conducting a comparative analysis of simulation results with existing routing protocols. The first section establishes that the best and most promising solution for organizing reliable communication between nodes in a UAV-based network is the use of a decentralized self-organizing network topology, called the Flying Ad Hoc Network (FANET). In the second section, the principles of the bee colony are discussed. The main characteristics of routing algorithms for self-organizing networks are given. The peculiarities of functioning of the developed routing algorithm on the basis of soy intelligence of the bee colony are described. The three-level structure of the self-organizing network node, which is used in the proposed algorithm, is presented and the characteristics of functioning of each level are described in detail, using the model of agents, created by analogy with wildlife, where each node acts as a hive, and the route is a source of food. In the third section the metrics, which affect the performance of the routing algorithm in FANET networks are considered and formulas for their calculation are given. The packet formats of the developed algorithm and the process of establishing a route, as well as data transmission and broadcast message propagation control are described. In the fourth section, the selection of indicators to evaluate the effectiveness of the routing process is carried out. A comparative analysis of the proposed routing algorithm with AODV and OLSR protocols was conducted. The results confirmed the effectiveness of the algorithm in self-organizing UAV networks. The proposed algorithm provided high network performance, surpassing the AODV and OLSR protocols in some respects. The fifth section provides an economic assessment of the feasibility of developing a software product based on swarm intelligence. Based on the calculations, it is shown that this project is cost-effective, and its development in Swift can significantly reduce implementation costs. Keywords: FANET; self-organized networks; routing; swarm intelligence; bees; routing metrics.