针对离线集中式卫星资源调度算法不能有效适应任务与资源动态变化和不能满足任务高实时性的问题，研究了未来天基激光网络中在线分布式的接入调度方法。分析了激光网络接入调度问题特点和约束条件，通过引入排队策略实现了算法的分布计算和在线调度；通过增加抢占优先权策略实现了网络对实时任务的支持；建立了接入节点上的排队系统模型，利用二维马尔可夫链对模型进行分析求解，得到了平均队长、平均等待时间和接入阻塞率等性能参数；给出了一个仿真场景，并对该算法进行数值仿真，结果表明，与离线集中式调度算法相比，在线分布式算法将任务等待时延降低了10%以上，阻塞率降低了90%以上，达到了实现任务在线调度和提高调度实时性与高效性的目的。

Because the off-line centralized satellite resource scheduling algorithm cannot adapt the dynamic variety of mission and resource and meet the high real time of the mission, an on-line distributed scheduling algorithm designed for the future space-based laser satellite network is studied. The characteristics and constraints of access scheduling are analyzed. The features of on-line scheduling and distributed computing of the algorithm are achieved by introducing the queuing scheme; preemptive priority is brought into schedule tasks in real time. Queuing system model in access node is established and a two-dimension Markov chain is used to solve and analyze the model, and some performance parameters, including average queue length, average waiting delay and blocking probability, are obtained. Through a simulated scene, the numerical simulation of the algorithm is done. The results show that the online distributed algorithm reduces waiting delay by more than 10% and blocking rate by 90% compared to off-line centralized algorithm, achieving the goal of on-line scheduling, real-time and efficiency.