基于小波的DVS在无线传感器网络中的应用  

Dynamic voltage scaling in wireless sensor network based on wavelet

作　　者：陈欢[1] 陈向东[1] 胡黎黎[1] 张传武[2] 

机构地区：[1]西南交通大学信息科学与技术学院,成都610031 [2]西南民族学院电气信息工程学院,成都610041

出　　处：《计算机应用研究》2007年第8期262-263,266,共3页Application Research of Computers

基　　金：国家自然科学基金资助项目(60572026)

摘　　要：为了最大限度节约能源的使用,使无线传感器网络使用寿命延长,针对无线传感器网络的节点在运行状态提出了基于小波的动态电压调度(DVS)算法。该方法首先通过使用小波、Kalman和AR三种算法相结合对下一任务量进行预测,并根据预测结果判断处理器(使用SA-1100处理器)进入何种操作频率等级(处理器的工作电压与操作频率一一对应),使较小任务量进入较小操作频率等级的状态,从而减少了节点在周期内的能量消耗。仿真实验结果表明,该方法是有效的。In order to extend life of wireless sensor network by making economical use of energy, put forward an effective method of dynamic voltage scaling which is based on wavelet, aimed at the node of wireless sensor network in running state to resolve it. Used wavelet, Kalman filter and AR to forecast the workload which next period will be, and what processors operating frequency levels（ the work voltage and the operating frequency of CPU are one by one） is judged to enter according to the forecast results, the smaller task, the smaller operating frequency using, thereby reducing the energy consumption of nodes in one period. Experimental results show that the method is effective.

关 键 词：无线传感器网络 动态电压调度 小波 卡尔曼 自回归 SA-1100处理器 

分 类 号：TP212.5[自动化与计算机技术—检测技术与自动化装置]