基于贝叶斯估计的无线传感器网络链路选择算法

Algorithm for probabilistic link selection in wireless sensor networks using Bayesian estimation

为了在能量受限的无线传感器网络中以较低的控制开销选择出具有高分组递交率的链路，提出一种基于贝叶斯估计的依概率链路选择算法（BPLS）．该算法将链路探测过程分成若干轮，在每轮中对链路分组递交率进行贝叶斯估计，依据估计结果决定下一轮探测中选择该链路的概率．在此基础上设计了可靠的路由算法．仿真结果表明：BPLS算法能够快速挑选出高质量链路；当控制开销较低时，选出质量最高链路的成功率比naive算法提高10％～20％；基于BPLS的树形路由在分组递交率和每分组能耗上优于基本的树形路由．

In order to select links with high packet delivery ratio at low control overhead in energy constrained wireless sensor networks, a Bayesian based probabilistic link selection （BPLS） algorithm is proposed. In this algorithm, link quality estimation is divided into rounds. At the end of each round, Bayesian estimation of link reliability is carried out based on the feedback information, and the estima- tion results are in turn used to determine link selection probability in next round. BPLS is incorporated into tree routing algorithm for reliable data collection. Simulation result shows that BPLS rapidly converges to the best link; at low control overhead it has a 10%-20% higher success rate than naive algorithm in selecting the best link; BPLS based tree routing algorithm outperforms basic tree algorithm in data packet delivery rate and energy consumption per packet.