Resource allocation and adaptive routing in multimedia low Earth orbit satellite mobile networks

The paper focuses on the issues of admission control and rerouting for LEO satellite networks providing multimedia services. The issues includes resource allocation and routing for intersatellite links (ISLs), load balance amongst links, and satellite handover and rerouting caused by satellites' dynamic characteristics. Applying Markov decision process (MDP) theory and generic policy-iteration formulation, a distributed routing scheme, called adaptive cost routing (ACR), is proposed. Comparing the performance of ACR with other routing schemes in terms of system throughput and blocking rate, the ACR presents not only a better routing performance, but also less connection setup time. It is also easier to implement than most centralized routing algorithms.