基于多级队列的LARS数据流调度算法的设计与仿真研究

Design and Simulation of LARS Data Stream Scheduling Algorithm Based on Multilevel Queue

文章提出了一种新的移动通信系统的数据流优化调度算法,即基于多级队列的LARS数据流调度算法(MLARS),并分析了其调度参数的设置。该算法利用多级队列技术,将待传输的移动通信数据流按到达时序分成多个队列,每个队列中优选出优先级最高的数据流并进行发送。与普通的LARS数据流调度算法不同的是,该算法适合于保证物联网数据的传输和发送,可有效防止饥饿现象和大数据包长期占用信道现象的发生。系统仿真分析的结果显示在用户数较小时,MLARS算法和LARS算法的性能差别不是很大,但是随着系统中用户数目增多,系统负荷变大时,采用MLARS算法的系统吞吐率比采用LARS算法的系统高12%,且MLARS算法能满足不同用户对服务质量的个性化要求,可广泛应用于LBS、传感器网络等物联网通信服务系统中。

This paper presents a new data flow optimization scheduling algorithm for mobile communication systems, which is the LARS data flow scheduling algorithm based on multilevel queue （MLARS）, and analyzes its scheduling parameters. In this algorithm, the mobile communication data stream is divided into multiple queues according to the arrival time sequence, and the highest priority is selected in each queue. The difference with the ordinary Lars data flow scheduling algorithm is, the algorithm is suitable to ensure data networking transmission and the delivery, it can effectively prevent occurrence of hunger and large packet long-term occupation of the channel of the phenomenon. System simulation results show when the number of users is small, MLARS algorithm and the LARS algorithm performance difference is not great, but with the increase in the number of users in the system, the system load larger, the system throughput MLARS algorithm rate than the Lars algorithm system of 12%, and algorithm MLARS can meet the requirements of different users on the quality of service of personalized, it can be widely used in LBS, sensor networks, network communication service system.