有向感知可充电传感器节点的动态激活策略

摘要

In energy harvesting directional sensor networks,node activation,sleeping scheduling and direction selection have a great influence on the target coverage and monitoring probability. Regarding the two cases of rechar-ging speed being slow than discharging speed and recharging speed being fast than discharging speed,sequential Greedy Node Activation Scheme(SGA)and sequential Greedy Node Inactivation Scheme(SGI)are proposed. SGA scheme sequentially allocates the node which can bring the maximal coverage probability gain to its corresponding slot and sensing direction.After choosing the optimal sensing direction for all the sensor nodes in all the slots,SGI scheme sequentially selects the nodes which cause the minimal system coverage loss one by one and inactivate them.Theoreti-cal analysis prove that SGA and SGI schemes guarantee at least 50% performance on system average coverage when compared with optimal activation scheme. Furthermore,they give a simple polynomial time solution to the NP-hard problem of maximizing system coverage. Finally,the performance of SGA and SGI schemes are verified by simulation.%在采集环境能量供电的有向感知传感器网络中,节点的激活、休眠调度及感知方向选择关系到对目标点的覆盖和监测成功率.基于节点采集能量速度慢于耗能速度及采集能量速度快于耗能速度两种情况,设计了逐次贪婪节点激活算法SGA和逐次贪婪节点休眠算法SGI.SGA算法逐次将能够给系统带来最大覆盖率增益的感知节点分配给其对应的时隙及感知方向;在为所有感知节点在所有时隙选择最佳感知方向激活的基础上,SGI算法逐次选择令系统覆盖率损失最小的感知节点,令其在相应的时隙休眠.理论证明,SGA算法及SGI算法至少能获得最优激活调度算法50%的系统平均覆盖率性能,且将最大化系统覆盖率的NP难问题转化为简单的多项式时间方法.最后,仿真验证了SGA算法和SGI算法的性能.