基于多关联模型的软件定义高密WLAN EDCA机制的吞吐量优化

Throughput Optimization of multiple association-based EDCA software defined dense WLAN

为了提高IEEE 802.11e增强分布式信道接入(EDCA)机制在高密场景下的全局吞吐量,研究了基于多关联模型的软件定义WLAN系统和无线接入点(AP)关联算法。首先扩展了软件定义WLAN架构,使之在不改变终端协议的基础上支持多关联,实现将同一终端的不同类型数据流由不同的AP进行传输。其次,考虑高密场景空口时间和链路速率特性,提出了快速分配算法,将算法的时间复杂度由2N简化到logN。最后实现了该机制并对其进行了验证,结果发现采用多关联模型的分配机制能明显提高EDCA机制下的全局带宽,相对于传统的单AP策略,全局带宽最高能提升48%;相对于高密AP全启用策略,全局带宽最高能提升35%。

To improve the global throughput of IEEE 802. 11 e enhanced distributed channel access（ EDCA） architecture under dense scenarios,the software defined WLAN system based on a multiple association model and a access point（ AP） association algorithm are studied. Firstly,the software defined WLAN architecture is extended to support multiple association without client-side modification,thus traffic in different priorities can be transmitted by different physical APs. Then the features of both airtime and data rate in dense scenarios are considered,and the quick AP selection algorithm is proposed. Its time complexity is decreased to logNfrom 2 N. Finally,the algorithm is implemented and evaluated in test-bed. The results show that the multiple association-based AP selection algorithm obviously improves the total throughput. Compared with the single AP selection policy,the total throughput is increased by 48%,while compared with the dense AP selection policy,the increasement is up to 35%.