RESEARCH ARTICLE Optimal routing strategy based on extreme learning machine with beetle antennae search algorithm for Low Earth Orbit satellite communication networks Aghila Rajagopal 1 | A. Ramachandran 2 | K. Shankar 3 | Manju Khari 4 | Sudan Jha 5 | Gyanendra Prasad Joshi 6 1 Department of IT, Sethu Institute of Technology, Kariapatti, Tamil Nadu, 626115, India 2 Department of Computer Science and Engineering, University College of Engineering, Panruti, Tamil Nadu, 607106, India 3 Department of Computer Applications, Alagappa University, Karaikudi, 630003, India 4 Computer Science Department, Ambedkar Institute of Technology, New Delhi, 110092, India 5 Department of Computer Science and Engineering, Chandigarh University, Mohali, Ludhiana, Punjab, 140413, India 6 Department of Computer Science and Engineering, Sejong University, 209, Neungdong-ro, Gwangjin-gu, Seoul, 05006, South Korea Correspondence Gyanendra Prasad Joshi, Department of Computer Science and Engineering, Sejong University, 209, Neungdong-ro, Gwangjin-gu, Seoul, 05006, South Korea. Email: joshi@sejong.ac.kr Summary Due to the significant utilization of terrestrial communication, Low Earth Orbit (LEO) satellite network is a critical part of satellite communication networks owing to its several benefits. But the efficient and trustworthy routing for LEO satellite networks (LSNs) is a difficult process because of dynamic topology, adequate link changes, and imbalanced communication load. This study devises a new hybridization of extreme learning machine (ELM) with multitask beetle antennae search (MBAS) algorithm- based distributed routing called the MBAS-ELM model. The proposed model deter- mines the routes based on traffic forecasting with respect to the level of traffic circu- lation on the earth. The proposed method is employed for traffic forecasting at the satellite nodes (SNs). To identify the optimal routes, mobile agents (MAs) are applied to concurrently and autonomously determine for LSNs and make a decision on rou- ting data. The experimental outcome has showcased the effective performance of the proposed model over the compared models in terms of different measures, namely, average delay, packet loss ratio (PLR), and queuing delay. The results are vali- dated under varying simulation time and data sensing rates. The obtained outcome pointed out the superior performance of the proposed MBAS-ELM model compared with other methods. KEYWORDS Low Earth Orbit, machine learning, mobile agents, routing, satellite networks 1 | INTRODUCTION In past decades, the widely used domains of speedy mobile Internet and fast development of space methodologies, satellite networks have been designed as an indivisible unit of the global mobile system. Due to worldwide coverage, satellite networks offer stable communication facilities for regions lacking terrestrial networks as Low Earth Orbit (LEO) satellite consists of lesser orbit altitude and comprises the merits of less transmit- ting latency as well as a loss of connectivity. Additionally, invulnerability is an alternate characteristic of the LEO satellite system because of the given reasons. Besides, the operational functions of LEO satellites are more reliable. Apart from this, the intersatellite connection creates an inter- action among satellites, which is autonomous from a terrestrial structure. Hence, LEO satellite communications capture maximum focus from the educational sector, companies, and other organizations. 1 The global coverage of the LEO satellite network requires 10–100 satellites due to relatively small coverage. Furthermore, it contains the fea- tures of a short orbit period, greater dynamic topology, and sequential link handovers (LHs). Also, an extended application of physical-layer Received: 25 June 2020 Revised: 1 November 2020 Accepted: 30 November 2020 DOI: 10.1002/sat.1391 Int J Satell Commun Network. 2020;1–13. wileyonlinelibrary.com/journal/sat © 2020 John Wiley & Sons, Ltd. 1