WPAN/WLAN应用环境下链路自适应的CIR预测算法

在分析了基于信干比（CIR）估计的链路自适应算法的基础上，指出该算法的性能取决于如何对下一调整间隔的CIR值进行预测，提出了CIR的线性预测算法，并在此基础上将自适应滤波理论引入到CIR的预测中，提出了适合于CIR预测的自适应算法，分析了此种情况下的收敛因子和其他参数的选取。数值仿真结果表明，自适应预测算法明显优于简单预测与线性预测算法，特别是在CIR估计存在误差时，优势更加明显，其吞吐量的最大增幅超过了5Mbit/s。需要指出的是，虽然本文的分析和仿真均以HIPERLAN2为例，但其结果对于在WPAN／WLAN应用环境下的其他系统也都适用。

Algorithm of adaptive prediction on CIR in link adaptation for WPAN/WLAN applications

Through the analysis of the performance of link adaptation based on carrier to interference ratio (CIR), a new adaptive prediction algorithm of CIR is proposed together with a linear prediction algorithm. The parameter selections of the algorithms are analyzed too. Simulation results indicate that the adaptive prediction algorithm is better than other algorithms in term of the network throughputs. Moreover, the maximum improvement on its throughputs can be achieved up to 5 Mbit/s. Although the algorithms are simulated in HIPERLAN2 network, they can also be extended into other applications such as WPAN/WLAN.