压缩感知下最短路径的无线网络数据收集算法  

A shortest path data collection algorithm for wireless networks under compressed sensing

作　　者：魏连锁[1] 马敬云 郭媛[1] WEI Liansuo;MA Jingyun;GUO Yuan(Department of Computer and Control Engineering,Qiqihar University,Qiqihar,Heilongjiang 161006,China)

机构地区：[1]齐齐哈尔大学计算机与控制工程学院,黑龙江齐齐哈尔161006

出　　处：《东北石油大学学报》2022年第3期98-106,I0007,I0008,共11页Journal of Northeast Petroleum University

基　　金：国家自然科学基金项目(61872204);黑龙江省自然科学基金项目(LH2019F037);黑龙江省教育厅面上项目(135409312)。

摘　　要：针对现有压缩感知(CS)数据收集方法存在能耗高、传输路径冗余等问题,为实现有效数据收集,提出一种基于稀疏随机投影压缩感知下最短路径的无线网络数据收集算法,分析算法能耗优势。引入可变概率将待收集数据分类实现稀疏投影,建立稀疏随机投影压缩感知模型;结合最短路径(Dijkstra)算法提出路由策略建立网络能耗模型;各节点根据自身收集信号强度选择传输概率发送数据,利用Sink节点收集到的数据重构。仿真结果表明,文中算法与对比算法相比,在保证原始数据重构精度的前提下,能够减少数据传输过程中的数据量,均衡网络能耗,达到延长传感器网络寿命的目标。In order to realize effective data collection,aiming at the problems of high energy consumption and redundancy of transmission path in existing compressed sensing(CS)data collection methods,this paper proposes a wireless network data collection algorithm based on the shortest path under sparse random projection compressed sensing.And the energy consumption advantage of the algorithm is analyzed.Variable probability is introduced to classify the data to be collected for sparse projection,and the compressed sensing model of sparse random projection is established.Combined with the shortest path(Dijkstra)algorithm,routing strategy is proposed to build a network energy consumption model.Each node selects transmission probability to send data according to its own signal intensity,and uses the data collected by the Sink node for reconstruction.Simulation results show that compared with the comparison algorithm,the proposed algorithm can reduce the amount of data in the process of data transmission,balance the energy consumption of the network,and prolong the life of the sensor network on the premise of ensuring the accuracy of original data reconstruction.

关 键 词：压缩感知 最短路径树 无线传感器网络 数据收集 能耗 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]