Agent-Based Load Balancing and Qos Routing for Leo Satellite Networks

This paper proposes and evaluates the Agent-based load balancing and QoS routing (ALBQR) in the low earth orbit (LEO) satellite networks. Two kinds of agents are designed. Stationary agents located at the satellites calculate link cost using the sum of propagation and queuing delays, employ FARIMA model to predict the traffic of satellite node, modify the link cost based on the predicted traffic. Mobile agents migrate among satellites to gather link status information, notify stationary agents to update routing tables and reserve resource. Mobile agents carry user's delay, bandwidth and packet loss rate, explore the routing path of satisfy the lowest path cost and QoS constraints. Through simulations on an Iridium satellite network, ALBQR is shown to achieve a balanced traffic distribution among satellites, provide better throughput, decrease packet loss and average end-to-end delay.