Dynamic Contact Plan Design in Broadband Satellite Networks With Varying Contact Capacity

Contact plan design (CPD) aims to schedule conflicting forthcoming communication opportunities (i.e., contacts). In CPD, the capacity of established contacts is indeed time-varying. Neglecting such variations in contact capacity inevitably leads to system performance deterioration. In this letter, we develop a new dynamic CPD algorithm while considering the time-varying property of contacts in broadband data relay satellite networks. The novelty lies in that we treat the flow optimization problem in a time-expanded graph as a queue-stability-related stochastic optimization problem, such that the problem can be solved independently slot by slot. Furthermore, we prove that the proposed online algorithm can approach arbitrarily close to the optimal throughput. Simulation results validate the efficiency of our proposed algorithm.