基于改进压缩感知的战勤网络状态监测模型

Logistic network monitoring model based on improved compressed sensing

为提高战勤网络信息感知节点状态的监测效率,对经典压缩感知理论进行了改进,构造了一种适合于测量战勤网络状态的行和为零的贝努利测量矩阵,理论证明了原始重构算法依然适用于改进后的目标函数,同时提出了基于压缩感知的战勤网络状态监测模型。将改进后的压缩感知方法应用于提出的检测模型,对Hadoop进度跟踪机制进行改进,并在仿真环境下对解码精度、压缩比率、定位效率进行了测试和分析。测试结果表明:在保持与传统方法相同的监测精度下,新方法有效检测的战勤网络规模可提高约16倍。

In order to improve the monitoring efficiency of the information sensing node of logistic network,the classical theory of compression perception was improved to construct a suitable service network for state measurement and measurement matrix for Bernoulli zero,whereby to prove that the original objective function reconstruction algorithm was still applicable to the improved logistic network state monitoring model compression.The Hadoop schedule tracking mechanism was improved by using the improved method of compressed sensing,and decoding accuracy,compression ratio and the positioning efficiency was tested and analyzed in the simulation environment.Test results show that by keeping the same precision with the traditional monitoring method,the new method can effectively detect the logistic network with the scale increase of about 16 times.