基于投影矢量的双组播树高效路由数据收集

现有的节点数据收集算法复杂度高，与路由结合效果不理想，且不能满足无线传感器网络高效能量的节点数据收集，而压缩感知理论具有容错性好、编码简单的优点。基于压缩感知的特性，提出了一种传感器网络中基于投影矢量的双组播树高效路由数据收集，该算法将贝叶斯压缩感知理论与传感器路由相结合，解决现有算法不能满足传感器对能耗敏感的问题。算法的基本思想首先根据初始观测矢量来寻求能量高效并得到合适路由的最优投影。然后利用节点系数能耗最小与广义矢量的主分量作为目标节点，采用微分嫡改变量最大的原则进行求解节点最佳投影系数，最后在sink与目标节点路由问题上采用正逆向组播树进行路由构造。理论和仿真结果表明在保证能耗的条件下取得了较好的重建仿真效果，对无线传感器通信具有一定的实用价值。

High-efficiency Routing Data Collection of Dual Multicast Tree Based on The Projection Vector

The present node data gathering algorithm is complex, and its routing binding effect is not ideal, and it cannot meet the high energy node data collection of wireless sensor network, whereas compressed sensing theory has the merits of fault tolerance and simple coding. Based on compressed sensing and its merits, The method that high-efficiency routing data collection of dual multicast tree based on the projection vector in sensor networks is proposed. It combining Bayes compressed sensing theory with sensor route, solves the problem of existing algorithm, that is, the sensor is sensitive to energy consumption. The basic idea of the algorithm is to firstly seek energy efficient and appropriate routing optimal projection according to the initial observation vector, and secondly use the node coefficient of minimum energy consumption and the principal component of generalized vector as the target node and the principle of maximum differential entropy change for node optimal projection coefficient, and lastly use reverse multicast routing structure in the problem of sink and the target node routing. Theoretical and simulation results indicate that it has obtained better reconstruction effect of simulation under the condition of ensuring energy consumption, which has a certain practical value on wireless sensor communication.