Dynamic bandwidth allocation for multimedia traffic in TDMA broadband satellite networks

Abstract

A number of satellite systems are currently under development, planning to offer a wide range of traffic services to a large and diverse number of users. In order for these to be financially competitive, the use of the satellite bandwidth, the most expensive component of these systems, needs to be maximized. In this paper we address two ways we can do this. We first discuss a dynamic channel allocation policy, more suitable for a network consisting of a large number of users with fairly limited and changing bandwidth demands. We then develop a Markov Decision Process that could be used to optimize the bandwidth allocation in a broadband TDMA environment and which provides an upper bound for the best allocation policy available. Since the implementation of a computationally intensive and complex optimization policy could prove impractical in a real system, we finally introduce some candidate sub-optimal allocation schemes, that might be less efficient than the optimal, but could prove a lot easier to implement.